JP5397177B2 - Content recommendation device, content recommendation program - Google Patents

Description

The present invention relates to a content recommendation device for recommending contents such as programs and music, but about the content recommendation program.

  In a content viewing apparatus for viewing content transmitted via broadcast or communication, a technique for inferring a user's preference and interest from a content viewing history, etc., and recommending content that matches the user's preference and interest is known. For example, in a program viewing device for viewing and recording a television program, a content recommendation device has been proposed that estimates a user's preference based on program information of the recorded program and recommends a program that matches the user's preference. (For example, refer to Patent Document 1).

JP 2008-205621 A

  However, in the content recommendation device as described in Patent Document 1, in order to estimate the user's preference based on the program information of the recorded program, the program recommended to be determined to match the user's preference is A program similar to a program recorded in the past, and a program similar to a program that matches the user's preference has not been broadcast in the past, or the user is not aware that it has been broadcast. There is a problem that if the program is not recorded, it is not determined that the program matches the user's preference.

Accordingly, the present invention is a content conforming to the tastes and interests of the user, the user content recommendation apparatus capable of recommending in content that is not to view content that is similar to the past, to provide a content recommendation program For the purpose.

  In order to achieve the above object, the first feature of the content recommendation device according to the present invention is that operation history storage means for storing operation history of the content recommendation device by one or more users, and preference information of the one or more users. A preference information collecting means for collecting the usage history, and an operation history of the one or more users is read from the operation history storage means, and a usage time histogram is calculated for each of the one or more users based on the read operation history. Time histogram calculation means; the usage time histogram included in the one or more user preference information collected by the preference information collection means; and the one or more user usage times calculated by the usage time histogram calculation means. A user of the usage time histogram included in the preference information and the usage time histogram A user identifying means for identifying whether or not the user of the usage time histogram calculated by the ram calculating means is the same; and if the user is identified by the user identifying means as being the same Based on the preference information integrated for each user identified to be the same by the preference information integration unit, the preference information integration unit that integrates the preference information for each identified user, Content preference level calculation means for calculating the content preference level for each user identified as the same, and the content preference level of the content scheduled to be broadcast in the future calculated for each user identified as the same Content preference storage means for performing content recommendation storage when requested by the user for content recommendation From reading the content preference of the content to be broadcast in the future that corresponds to the user who the operation is to and a content recommendation unit for recommending the content based on the content preference.

  In order to achieve the above object, a second feature of the content recommendation device according to the present invention is that operation history storage means for storing operation history of the content recommendation device by one or more users, and preference information of the one or more users. A preference information collecting means for collecting the usage history, and an operation history of the one or more users is read from the operation history storage means, and a usage time histogram is calculated for each of the one or more users based on the read operation history. Time histogram calculation means; the usage time histogram included in the one or more user preference information collected by the preference information collection means; and the one or more user usage times calculated by the usage time histogram calculation means. A user of the usage time histogram included in the preference information and the usage time histogram A user identifying means for identifying whether or not the user of the usage time histogram calculated by the ram calculating means is the same; and if the user is identified by the user identifying means as being the same The preference information integration unit that integrates the preference information for each identified user, and the keyword preference level included in the preference information for each user identified as the same integrated by the preference information integration unit When the user's operation requests content recommendation, the keyword preference storage unit reads out the keyword preference corresponding to the operated user from the keyword preference storage, and is included in the read keyword preference Search results sent from the server device. Receiving a content list is to and a content recommendation unit that outputs the received content list as a content recommendation screen.

  In order to achieve the above object, the third feature of the content recommendation device according to the present invention is that operation history storage means for storing operation history of the content recommendation device by one or more users, and content storage means for storing content information. Reading out the content information from the content storage means, extracting a keyword from the read content information, and reading out the operation history of the one or more users from the operation history storage means A usage time histogram calculating means for calculating a usage time histogram for each of the one or more users based on the operated history, the keyword extracted by the keyword extraction means, and the read from the operation history storage means Key for calculating the preference degree of the keyword based on the operation history A preference degree calculation means, a set of the keyword and the preference degree of the keyword calculated by the keyword preference degree calculation means, and a usage time histogram for each of the one or more users calculated by the usage time histogram calculation means Are stored for each of the one or more users by keyword preference storage means, preference information collection means for collecting preference information of the one or more users, and the one or more collected by the preference information collection means The usage time histogram included in the user's preference information and the usage time histogram of the one or more users calculated by the usage time histogram calculation means, and the user of the usage time histogram included in the preference information , The usage time histogram calculated by the usage time histogram calculation means If the user identification means for identifying whether the user is the same or the user identification means is identified by the user identification means, the keyword and the keyword preference degree from the keyword preference storage means A preference information integration unit that reads a set, and for each user identified as the same by the user identification unit, integrates the read keyword and the keyword preference set into the preference information; Content preference for calculating the content preference level of the content scheduled to be broadcast in the future for each user identified as the same based on the preference information integrated for each user identified as the same by the preference information integration means The content preference of the content scheduled to be broadcast in the future calculated for each user identified to be the same as the degree calculation means Content preference storage means for storing the preference, and when content recommendation is requested by the user's operation, the content preference of the content scheduled for future broadcast corresponding to the operated user from the content information storage means And content recommendation means for recommending the content based on the content preference level.

  In order to achieve the above object, a first feature of the content recommendation program according to the present invention is an operation history storage step of storing an operation history of the content recommendation device by one or more users in an operation history storage means; A preference information collecting step for collecting user preference information, and reading the operation history of the one or more users from the operation history storage means, and using time for each of the one or more users based on the read operation history A usage time histogram calculating step for calculating a histogram; the usage time histogram included in the one or more user preference information collected by the preference information collecting step; and the one or more calculated by the usage time histogram calculation step. The usage time histogram is compared with the usage time histogram of the user and the usage information included in the preference information The user identification step for identifying whether or not the user of the inter-histogram and the user of the usage time histogram calculated by the usage time histogram calculation step are the same, and the user identification step identifies that the user is the same The preference information integration step for integrating the preference information for each user identified as the same, and the preference information integrated for each user identified as the same in the preference information integration step. A content preference level calculating step for calculating a content preference level of content scheduled to be broadcast in the future for each user identified as the same, and a future broadcast schedule calculated for each user identified as the same A content preference storage means for storing the content preference level of the content of When content recommendation is requested by the user's operation in the preference storage step, the content preference of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage means, and the content preference And causing the computer to execute a content recommendation step of recommending the content based on the degree.

  In order to achieve the above object, a second feature of the content recommendation program according to the present invention is that an operation history storage step of storing an operation history of the content recommendation device by one or more users in an operation history storage means; A preference information collecting step for collecting user preference information, and reading the operation history of the one or more users from the operation history storage means, and using time for each of the one or more users based on the read operation history A usage time histogram calculating step for calculating a histogram; the usage time histogram included in the one or more user preference information collected by the preference information collecting step; and the one or more calculated by the usage time histogram calculation step. The usage time histogram is compared with the usage time histogram of the user and the usage information included in the preference information The user identification step for identifying whether or not the user of the inter-histogram and the user of the usage time histogram calculated by the usage time histogram calculation step are the same, and the user identification step identifies that the user is the same The preference information integration step for integrating the preference information for each user identified as the same, and the preference for each user identified as the same integrated in the preference information integration step. The keyword preference storage step for storing the keyword preference included in the information in the keyword preference storage unit, and when the content recommendation is requested by the user operation, the keyword preference storage unit corresponds to the operated user. The keyword preference level is read out, and the read keyword preference is read out A content recommendation step for requesting the server device to perform a search based on a keyword included in the server, receiving a content list as a search result transmitted from the server device, and outputting the received content list as a content recommendation screen. It is to be executed.

  In order to achieve the above object, a third feature of the content recommendation program according to the present invention includes an operation history storage step of storing operation history of the content recommendation device by one or more users in the operation history storage means, and content information. A content storing step to be stored in a content storing unit; a keyword extracting step of reading out the content information from the content storing unit; and extracting a keyword from the read content information; and the one or more users from the operation history storing unit A usage time histogram calculation step of calculating a usage time histogram for each of the one or more users based on the read operation history, the keyword extracted by the keyword extraction step, and the operation Read from history storage means A keyword preference calculation step for calculating a preference degree of the keyword based on the operated history, a pair of the keyword preference degree calculated by the keyword and the keyword preference calculation step, and the usage time histogram A keyword preference storage step of assigning the usage time histogram for each of the one or more users calculated in the calculation step to each of the one or more users and storing it in a keyword preference storage means; and the preference of the one or more users A preference information collecting step for collecting information, the usage time histogram included in the one or more user preference information collected in the preference information collection step, and the one or more calculated in the usage time histogram calculation step. Compare with user usage histogram, A user identification step for identifying whether or not the user of the usage time histogram included in the good information is the same as the user of the usage time histogram calculated by the usage time histogram calculation step, and the user identification step When the user is identified as the same, the keyword and the keyword preference set are read out from the keyword preference storage unit, and the read out is performed for each user identified as the same in the user identification step. The preference information integration step of integrating the keyword and the preference level of the keyword into the preference information, and the preference information integrated for each user identified as the same by the preference information integration step. Based on the content preference level of the content scheduled to be broadcast in the future. Then, the content preference calculation step for calculating for each identified user and the content preference for the content scheduled to be broadcast in the future calculated for each user identified as the same are stored in the content preference storage means. When content recommendation is requested by the content preference storage step and the user's operation, the content preference of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage means, and the content A content recommendation step of recommending the content based on the degree of preference is to be executed by a computer.

Content recommendation device according to the present invention, according to the content recommendation program, can be a content conforming to the tastes and interests of the user, to recommend content that is not viewing the content by the user is similar in the past.

It is a figure which shows the example of a connection with the content recommendation apparatus which is Example 1 of this invention, a preference information transmission apparatus, and a server apparatus. It is the figure which showed the structure of the content recommendation apparatus of Example 1 of this invention. It is a flowchart which shows the main processes for 1 period among the preference analysis processes which operate | move regularly by the predetermined time interval in the content recommendation apparatus of Example 1 of this invention. It is the figure which showed the structure of the preference information transmitted from a preference information transmission apparatus. It is the figure which showed an example of the usage time histogram. It is the figure which showed an example of the keyword preference degree contained in a preference information group. It is the figure which showed an example of program information. It is the figure which showed the structure of the content recommendation apparatus of Example 2 of this invention. In the content recommendation apparatus of Example 2 of this invention, it is a flowchart which shows the main processes for 1 period among the preference analysis processes currently operate | moving regularly with a predetermined time interval. It is a flowchart which shows the music recommendation process in the music recommendation part of the content recommendation apparatus of Example 2 of this invention. It is the figure which showed the structure of the preference information transmission apparatus of Example 3 of this invention. It is a flowchart which shows the main processes for 1 period among the preference analysis processes which operate | move regularly by the predetermined time interval in the preference information transmission apparatus of Example 3 of this invention. It is the figure which showed an example of keyword preference degree calculation. It is the figure which showed the structure of the preference information transmission apparatus of Example 4 of this invention. It is a flowchart which shows the main processes for 1 period among the preference analysis processes which operate | move regularly by the predetermined time interval in the preference information transmission apparatus of Example 4 of this invention. It is the figure which showed the connection structure of the content recommendation apparatus which is Example 5 of this invention, and a server apparatus. It is the figure which showed the structure of the content recommendation apparatus of Example 5 of this invention. In the content recommendation apparatus of Example 5 of this invention, it is a flowchart which shows the main processes for 1 period among the preference analysis processes currently operate | moving regularly with a predetermined time interval.

  Hereinafter, the form for implementing this invention is demonstrated in detail, showing an Example.

  In Embodiment 1 of the present invention, a content recommendation device that is connected to a preference information transmission device via a home network and recommends content that matches the user's preference based on the preference information received from the preference information transmission device is taken as an example. I will explain.

  FIG. 1 is a diagram illustrating a connection example of a content recommendation device 2, a preference information transmission device 8, and a server device that are Embodiment 1 of the present invention.

  In FIG. 1, the content recommendation device 2 and the plurality of preference information transmission devices 8 are connected via a home network 3, the home network 3 is connected to an external network, and the server device 5 is connected via the external network 4. Connected.

  The content recommendation device 2 has a function as a content viewing device for receiving and viewing content such as a television program transmitted via a broadcast wave and a video or music transmitted from the server device 5. The user's preference is estimated based on the preference information received from the preference information transmitting device 8 via the home network 3, and content that matches the user's preference is displayed on a display or the like. In the first embodiment, the content is described as a television program transmitted via a broadcast wave, for example.

  The preference information transmitting device 8 has a function as a content viewing device for receiving and viewing content such as a television program transmitted via a broadcast wave, a moving image or music transmitted from the server device 5, and the like. Further, preference information is generated based on a history of operations on the user's preference information transmission device 8, and preference information is transmitted in response to a request from the content recommendation device 2 connected via the home network 3. .

  The server device 5 is a server that distributes content such as moving images and music to clients, and responds to requests from the content recommendation device 2 and the preference information transmission device 8 that are clients via the home network 3 and the outside network 4. Content is distributed. Further, when a search is requested together with a search keyword from the content recommendation device 2 and the preference information transmission device 8, a content that matches the search keyword, that is, a list of content including the search keyword in information related to the content is transmitted as a search result. is there.

  In FIG. 1, the server device 5 may be omitted. In this case, the content recommendation device 2 and the preference information transmission device 8 view content such as TV programs transmitted via broadcast waves, and content such as videos and music recorded on physical media such as CDs and DVDs. It will be a thing.

<Configuration 1 of content recommendation device>
FIG. 2 is a diagram illustrating a configuration example of the content recommendation device according to the first embodiment of the present invention. The content recommendation device 2 shown in FIG. 2 has a function as a television recording device for viewing and recording a television program received via a broadcast wave.

  2, the content recommendation device 2 includes an input unit 201, an output unit 202, a broadcast receiving unit 203, a device control unit 204, an output control unit 205, a program recommendation unit 206, a recording unit 207, a program An information storage unit 208, a recording reservation information storage unit 209, a video / audio signal recording unit 210, an operation history storage unit 211, a program preference storage unit 212, a network connection unit 213, a preference information reception unit 214, A preference analysis unit 215.

  The input unit 201 includes an input device such as a remote controller, and an operation for instructing basic operation signals and content recommendation as a television recording device such as channel switching, recording reservation, and playback of a recorded program by a user operation. A signal is generated and supplied to the device control unit 204.

  The output unit 202 includes an image output device such as a liquid crystal display and an audio output device such as a speaker, and outputs video and audio supplied from the output control unit 205.

  The broadcast receiving unit 203 includes a tuner and a demodulating / separating unit, and from the device control unit 204 or the recording unit 207 among broadcast waves such as a terrestrial wave transmitted from a broadcast station and a satellite broadcast wave transmitted via a broadcast satellite. The broadcast wave of the designated channel is received, separated into a video / audio signal and program information, the video / audio signal is supplied to the output control unit 205 and the recording unit 207, and the program information is supplied to the program information storage unit 208. is there.

  Here, the program information is information added to the video / audio signal by the broadcasting station, and includes “title”, “channel”, “broadcast start time”, “broadcast end time”, “genre”, “content description” ”And“ Performer List ”. The “broadcast start time” and “broadcast end time” may include date information.

  The device control unit 204 sets a channel received from the broadcast receiving unit 203 based on the operation signal received from the input unit 201, stores recording reservation information in the recording reservation information storage unit 209, and records video and audio signals. The reproduction of the video / audio signal recorded in the unit 210 is started, and either the video / audio signal supplied from the broadcast receiving unit 203 or the video / audio signal read out from the video / audio signal recording unit 210 is output to the output control unit 205. Instructs whether to output to the unit 202.

  Here, the recording reservation information is information for executing recording by a timer, and includes information such as a recording channel and recording start / end times. Furthermore, the information of the folder for storing the recorded program is also included. A folder is for a user to organize recorded programs. Particularly when a plurality of users use a TV recording device, a folder corresponding to each user is prepared, and a user selects a favorite program. It records in its own folder.

  In addition, the device control unit 204 causes the operation history storage unit 211 to store information on the recording reservation operation by the user as an operation history. The operation history includes the time when the operation was performed and the recording reservation information, that is, the channel to be recorded, the recording start / end time, the folder information, and the like. The operation histories are sequentially stored, but since the storage capacity is limited, the old operation histories may be deleted.

  Further, the device control unit 204 instructs the program recommendation unit 206 to recommend a program together with folder information based on the operation signal received from the input unit 201, and the program supplied from the program recommendation unit 206 to the output control unit 205. Instructs the output of the recommendation screen.

  Based on an instruction from the device control unit 204, the output control unit 205 is a video / audio signal supplied from the broadcast receiving unit 203, a video / audio signal read out from the video / audio signal recording unit 210, and a program supplied from the program recommendation unit 206. One of the recommendation screens is output to the output unit 202.

  When the recommendation of the program is instructed together with the folder information from the device control unit 204, the program recommendation unit 206 reads the program information from the program information storage unit 208, and the program corresponding to the folder information instructed from the device control unit 204 The preference level is read from the program preference level storage unit 212, a predetermined number of programs are selected from the programs to be broadcast in the future, and the program information list of the selected program is output as a program recommendation screen. 202 is supplied.

  The recording unit 207 reads the recording reservation information from the recording reservation information storage unit 209 at a predetermined time period such as 1 minute, and executes the recording reservation when there is a recording reservation whose recording start time has arrived. The broadcast receiving unit 203 is instructed to receive the channel to be recorded, and the video / audio signal supplied from the broadcast receiving unit 203 is recorded in the folder included in the recording reservation information of the video / audio signal recording unit 210. In addition, when the recording end time of the recording reservation being executed comes, the recording reservation being executed is ended.

  The program information storage unit 208 is configured by a storage device such as a hard disk, for example, and stores program information supplied from the broadcast receiving unit 203.

  The recording reservation information storage unit 209 is configured by a storage device such as a hard disk, for example, and stores the recording reservation information supplied from the device control unit 204.

  The video / audio signal recording unit 210 is configured by a storage device such as a hard disk, for example, and assigns and records the video / audio signal supplied from the recording unit 207 to a folder instructed by the recording unit 207.

  The operation history storage unit 211 is configured by a storage device such as a hard disk, for example, and stores an operation history supplied from the device control unit 204.

  The program preference storage unit 212 is configured by a storage device such as a hard disk, for example, and stores a set of program identification information for uniquely identifying a program and a program preference assigned to a folder.

  The network connection unit 213 is a communication interface such as a network card, and communicates with the preference information transmitting device 8 connected to the home network 3 by connecting to the home network 3 through the network connection unit 213.

  When receiving preference information from the preference analysis unit 215, the preference information reception unit 214 communicates with the preference information transmission device 8 connected to the home network 3 via the network connection unit 213, and the preference information transmission device 8 Is requested to transmit preference information, the preference information transmitted from the preference information transmitting device 8 is received, and the received preference information is supplied to the preference analysis unit 215.

  The preference analysis unit 215 performs preference analysis processing at predetermined time intervals such as one day, one hour, one week, and the like. For example, the preference analysis unit 215 includes a preference information collection unit 215a and a usage time histogram. A calculation unit 215b, a user identification unit 215c, a preference information integration unit 215d, and a program preference level calculation unit 215e are provided.

  The preference information collection unit 215a instructs the preference information reception unit 214 to receive preference information and supplies the preference information supplied from the preference information reception unit 214 to the user identification unit 215c.

  The usage time histogram calculation unit 215b reads an operation history from the operation history storage unit 211, calculates a usage time histogram for each user folder, and supplies the calculated usage time histogram to the user identification unit 215c.

  The user identification unit 215c compares the usage time histogram included in the preference information supplied from the preference information collection unit 215a with the usage time histogram supplied from the usage time histogram calculation unit 215b, and collects preference information for each folder. The user of the preference information supplied from the unit 215a identifies whether or not the user of the preference information is the same as the user of the folder.

  The preference information integration unit 215d integrates the preference information for each folder of the user identified as the same when the user identification unit 215c identifies that the user is the same as the folder user, and the integrated preference information is displayed in the program. This is supplied to the preference level calculation unit 215e.

  The program preference calculation unit 215e reads program information of a program scheduled to be broadcast in the future from the program information storage unit 208, and a program scheduled to be broadcast in the future based on the preference information integrated for each folder supplied from the preference information integration unit 215d. Is calculated for each folder. Further, a set of the program identification information and the calculated program preference level is stored in the program preference level storage unit 212 for each folder.

  The content recommendation device 2 according to the first embodiment of the present invention includes a program information storage unit 208, a recording reservation information storage unit 209, a video / audio signal recording unit 210, an operation history storage unit 211, and a program preference storage. Although the unit 212 is configured as a separate storage device, each storage unit may be partially shared, or all the storage units may be shared by a single storage device.

  In addition, when one or two or more network compatible devices other than the content recommendation device 2 are connected to the home network 3, preference information from which the content recommendation device 2 receives preference information from among these network compatible devices. When the transmission device 8 is selected, the user of the content recommendation device 2 may select the preference information transmission device 8 by specifying the IP address or the like, or based on the UPnP (Universal Plug and Play) protocol. Thus, the preference information transmitting device 8 may be automatically discovered.

<Operation of Content Recommendation Device 2>
Next, the preference analysis process of the content recommendation device 2 that is Embodiment 1 of the present invention will be described in detail.

≪Preference analysis process≫
Details of the preference analysis processing in the preference analysis unit 215 of the content recommendation device 2 that is Embodiment 1 of the present invention will be described.

  FIG. 3 is a flowchart showing the main processing for one cycle of the preference analysis processing that is periodically operating at a predetermined time interval in the content recommendation device 2 that is Embodiment 1 of the present invention. .

  The preference information collection unit 215a of the preference analysis unit 215 receives preference information from the preference information transmission device 8 when the preference analysis processing is started (step S301). Specifically, the preference information collection unit 215a instructs the preference information reception unit 214 to receive preference information and receives the preference information supplied from the preference information reception unit 214.

  Here, the configuration of the preference information will be described.

  FIG. 4 is a diagram illustrating a configuration example of the preference information 400 transmitted from the preference information transmission device 8.

  The preference information 400 includes a plurality of preference information groups 401. Note that the preference information group 401 is created for each predetermined unit such as each user or each folder.

  Further, each preference information group 401 includes a usage time histogram 402 and a plurality of keyword preference levels 403.

  Furthermore, the usage time histogram 402 is a histogram of the usage frequency of the preference information transmitting apparatus for each time zone obtained by dividing 24 hours a day at intervals of, for example, 2 hours, and includes a plurality of time zones 404 and frequency 405 of the histogram. And a pair.

  On the other hand, the keyword preference degree 403 includes a keyword 406 and its preference degree 407.

  In step S301, the preference information 400 is received from each of the preference information transmitting devices 8 one by one.

  Next, the usage time histogram calculation unit 215b of the preference analysis unit 215 reads the operation history from the operation history storage unit 211 (step S302).

  Next, the number of folders prepared in the content recommendation device 2 is substituted into the variable N, and 0 is substituted into the variable i (step S303).

  Next, it is determined whether or not the value of the variable i is smaller than the value of the variable N (step S304). If it is determined that the value is smaller than the value of the variable N, the process proceeds to step S305. If it is determined that the value is not smaller than the value of the variable N, that is, it is equal to or greater than the value of the variable N, the preference analysis process ends.

  If it is determined in step S304 that the value of variable i is smaller than the value of variable N, 1 is added to the value of variable i and substituted for variable i (step S305).

  Next, the usage time histogram calculation unit 215b of the preference analysis unit 215 calculates a usage time histogram of the folder (step S306). Specifically, among the operation histories read out in step S302, for the operation histories in which the folder included in the operation histories is the i-th folder, for each time period in which 24 hours per day are divided at intervals of, for example, 2 hours. Then, the number of operation histories whose operation times are included in the time zone is counted, and a histogram is obtained in which the number of operation histories is the frequency of the time zone.

  Next, the user identification unit 215c of the preference analysis unit 215 calculates the usage time similarity of each preference information group 401 included in the preference information received in step S301 (step S307). Specifically, each of the usage time histogram 402 included in the preference information group 401 and the usage time histogram of the folder calculated in step S306 is represented by a multidimensional vector having a time zone as a dimension and a frequency as a value, The cosine distance between the two vectors is obtained, and the obtained cosine distance is set as the usage time similarity of the preference information group 401.

  Next, the user identification unit 215c of the preference analysis unit 215 selects the preference information group 401 whose usage time similarity obtained in step S307 is equal to or greater than a predetermined threshold (step S308). The preference information received in step S301 may include a preference information group 401 that reflects the preference of a user different from the user of the i-th folder, but the usage is similar to the usage time histogram of the i-th folder. It is considered that the preference information group 401 having a time histogram is likely to reflect the same user preference as the user of the i-th folder, and this step allows the same user as the user of the i-th folder. It is possible to select only the preference information group 401 that reflects the preference of the user. The threshold used here is set to an optimum value by experiment.

  Next, the preference information integration unit 215d of the preference analysis unit 215 integrates the plurality of preference information groups 401 selected in step S308 (step S309). Specifically, when there are a plurality of keyword preference levels 403 having the same keyword 406 for the keyword preference levels 403 included in the preference information group 401, the keywords 406 are added by adding the preference levels 407 of the keyword preference levels 403. The new preference degree 407 is assumed.

  Next, the program preference calculation unit 215e of the preference analysis unit 215 reads the program information of the program scheduled to be broadcasted from the program information storage unit 208 (step S310), and the program scheduled to be broadcasted from which the program information was read in step S310. The program preference level is calculated for each of these, and a set of the program identification information and the program preference level is assigned to the i-th folder and stored in the program preference level storage unit 212 (step S311). The program preference level calculation unit 215e calculates the program preference level, specifically, for each keyword preference level 403 included in the preference information group 401 integrated in step S309, the keyword 406 is the “title” of the program information, It is checked whether it is included in the “genre”, “content description”, “performer list”, etc., and the program preference level of the program is added to the preference level 407 of the keyword preference level 403 including the keyword 406 in the program information. And After executing this step, the process proceeds to step S304.

  Next, the calculation of the usage time similarity in step S307 shown in FIG. 3 will be described in detail using a specific example.

  FIG. 5 is a table showing an example of a usage time histogram.

  The usage time histogram example 12 shown in FIG. 5 shows specific examples of three types of usage time histograms when 24 hours per day are divided at intervals of 2 hours.

  The row of time zone 1202 indicates a time zone every two hours. For example, “0-2” indicates a time zone from 0:00 to 2:00.

  Specific examples A1203 to C1205 of usage time histograms are specific examples of three types of usage time histograms, and indicate the frequency for each time zone shown in the row of the time zone 1202, that is, the number of operation histories. For example, the frequency in the time zone from 18:00 to 20:00 in the specific example A1203 is “3”.

  The row with the number 1201 is assigned with a convenient number from 0 to 11 in each time zone.

  Next, a method for calculating the cosine distance will be described.

For the n-dimensional vector Va shown in (Formula 1) and the n-dimensional vector Vb shown in (Formula 2), the cosine distance cos θ between the vector Va and the vector Vb is calculated by the formula shown in (Formula 3). Here, Va · Vb indicates the inner product of the vector Va and the vector Vb, and is calculated by the equation shown in (Equation 4). || Va || indicates the magnitude of the vector Va, and is calculated by the equation shown in (Equation 5). || Vb || indicates the magnitude of the vector Vb, and is calculated by the same equation as (Equation 5). Equations 1 to 5 are as follows.

  The usage time similarity in the usage time histogram example 12 is a cosine distance cos θ shown in (Expression 3) when the usage time histogram is a 12-dimensional vector. When calculated according to this mathematical formula, for example, the usage time similarity between the specific example A1203 and the specific example B1204 is about 0.97. Further, the usage time similarity between the specific example B1204 and the specific example C1205 is about 0.07. The usage time similarity between the specific example C1205 and the specific example A1203 is about 0.03.

  Next, the calculation of the program preference in step S311 will be described in detail using a specific example.

  FIG. 6 is a diagram illustrating an example of keyword preference levels included in the preference information group integrated in step S309.

  In the keyword preference 13 shown in FIG. 6, the keyword column 1301 indicates specific keywords. Further, the preference level column 1302 indicates the value of the preference level corresponding to each keyword in the keyword column 1301.

  FIG. 7 is a diagram showing an example of program information for one program read in step S310.

  In FIG. 7, the program information 14 includes information such as a title 1401, a genre 1402, a content description 1403, and a performer list 1404.

  Here, the program information 14 includes keywords such as “world heritage”, “documentary”, and “gourmet” among the keywords included in the keyword preference degree 13. Therefore, the program preference level of this program is obtained as 1.14 by adding the preference levels of these keywords.

  Then, in the preference analysis unit 215 of the content recommendation device 2 shown in FIG. 3, the folder corresponding to the user who has operated from the device control unit 204 after completing the preference analysis processing that operates periodically at a predetermined time interval. When the recommendation of the program is instructed together with the information, the program recommendation unit 206 reads the program information from the program information storage unit 208, and determines the program preference level corresponding to the folder information instructed from the device control unit 204 as the program preference level. A predetermined number of programs that are read from the storage unit 212 and are broadcast in the future are selected from the programs with the highest program preference, and a list of program information of the selected programs is supplied to the output unit 202 as a program recommendation screen. Recommend the program that is the content to the corresponding user.

  Therefore, according to the content recommendation device 2 of the first embodiment, the operation history of the device by a plurality of users is stored, the preference information of the plurality of users is collected, the operation history of the plurality of users is read, and the plurality of users The usage time histogram is calculated for each folder, the usage time histogram included in the collected preference information of the plurality of users is compared with the usage time histogram of the plurality of users, and the user of the preference information is calculated for each user folder. Is integrated with the preference information identified as being the same user as the user of each folder for each folder of the plurality of identified users. Based on the integrated preference information, the program preference level of the program scheduled to be broadcast in the future is calculated for each folder, and the program preference level of the program scheduled to be broadcast in the future is calculated for each folder. Stores and reads the program preference of a program scheduled to be broadcast in the future, selects a predetermined number of programs from the programs that will be broadcast in the future, and selects a predetermined number of programs, and recommends a list of program information for the selected program Since the screen is recommended, it is possible to recommend a program that matches the user's preference and interest and that the user has not viewed a similar program in the past.

  Next, an example of a content recommendation device according to the second embodiment of the present invention will be described. In the second embodiment, an example in which music is used as content will be described.

<Configuration 2 of content recommendation device>
FIG. 8 is a diagram illustrating a configuration example of the content recommendation device according to the second embodiment of the present invention. The content recommendation device 2A shown in FIG. 8 has a function as a music playback device that plays back music received from the server device 5.

  The content recommendation device 2A includes an input unit 501, an output unit 202, a music reception unit 503, a device control unit 504, an output control unit 505, a music recommendation unit 506, an audio signal recording unit 510, and an operation history storage. A unit 511, a keyword preference storage unit 512, a network connection unit 213, a preference information reception unit 214, and a preference analysis unit 515.

  Here, among the configurations provided in the content recommendation device 2A according to the second embodiment of the present invention, the output unit 202, the network connection unit 213, and the preference information reception unit 214 are respectively the content recommendation device 2 according to the first embodiment. Since the configuration is the same as that in FIG.

  The input unit 501 includes an input device such as a remote controller, and a basic operation signal as a music playback device such as music playback by user operation, music reception from the server device 5, and an operation signal instructing music recommendation. Is generated and supplied to the device control unit 504.

  The music receiving unit 503, when instructed to receive music from the device control unit 504, communicates with the server device 5 connected to the outside network 4 via the network connection unit 213 via the home network 3, and the server The apparatus 5 is requested to transmit music, and the audio signal of the music transmitted from the server apparatus 5 is received and recorded in the audio signal recording unit 510.

  The device control unit 504 instructs the music reception unit 503 to receive music based on the operation signal received from the input unit 501 and starts playing the audio signal recorded in the audio signal recording unit 510. The output control unit 505 is instructed to output the audio signal read from the audio signal recording unit 510 to the output unit 202.

  In addition, the device control unit 504 stores information on the music reproduction operation by the user in the operation history storage unit 511 as an operation history. The operation history includes time when the operation is performed, music identification information for identifying the music, and the like.

  Also, the device control unit 504 instructs the music recommendation unit 506 to recommend music based on the operation signal received from the input unit 501, and outputs the music recommendation screen supplied from the music recommendation unit 506 to the output control unit 505. Is instructed.

  The output control unit 505 causes the output unit 202 to output either an audio signal read from the audio signal recording unit 510 or a music recommendation screen supplied from the music recommendation unit 506 based on an instruction from the device control unit 5204. is there.

  When music recommendation is instructed from the device control unit 504, the music recommendation unit 506 reads the keyword preference from the keyword preference storage unit 512 and is connected to the outside network 4 via the network connection unit 213. 5 via the home network 3, requesting the server device 5 to search for a keyword included in the read keyword preference degree, and receiving and receiving a music list as a search result transmitted from the server device 5. The music list is supplied to the output control unit 505 as a music recommendation screen.

  The audio signal recording unit 510 is constituted by a storage device such as a hard disk, for example, and records the audio signal supplied from the music receiving unit 503.

  The operation history storage unit 511 is configured by a storage device such as a hard disk, for example, and stores an operation history supplied from the device control unit 504.

  The keyword preference storage unit 512 is configured by a storage device such as a hard disk, for example, and stores a set of keywords and preference levels in association with each other.

  The preference analysis unit 515 executes preference analysis processing at a predetermined time interval such as one day, one hour, one week, etc. For example, the preference analysis unit 515 includes a preference information collection unit 215a and a usage time histogram. A calculation unit 515b, a user identification unit 515c, and a preference information integration unit 515d are provided.

  Here, among the configurations provided in the preference analysis unit 515, the preference information collection unit 215a is the same as the configuration with the same reference numerals provided in the preference analysis unit 215 of the content recommendation device 2 of the first embodiment. Therefore, explanation is omitted.

  The usage time histogram calculation unit 515b reads an operation history from the operation history storage unit 511, calculates a usage time histogram, and supplies the calculated usage time histogram to the user identification unit 515c.

  The user identification unit 515c compares the usage time histogram included in the preference information supplied from the preference information collection unit 215a with the usage time histogram supplied from the usage time histogram calculation unit 515b, and supplies it from the preference information collection unit 215a. This is to identify whether or not the user of the preference information is the same as the user of the content recommendation device 2A.

  The preference information integration unit 515d integrates the preference information identified as being the same user as the user of the content recommendation device 2A, and stores the keyword preference level included in the integrated preference information in the keyword preference level storage unit 512. Is.

  In the content recommendation device 2A according to the second embodiment of the present invention, the audio signal recording unit 510, the operation history storage unit 511, and the keyword preference storage unit 512 are configured as separate storage devices. The storage unit may be partially shared, or all the storage units may be shared by a single storage device.

  In addition, when one or two or more network compatible devices other than the content recommendation device 2A are connected to the home network 3, preference information from which the content recommendation device 2A receives preference information from among these network compatible devices. When the transmission device 8 is selected, the user of the content recommendation device 2A may select the preference information transmission device 8 by specifying the IP address or the like, or based on the UPnP (Universal Plug and Play) protocol. Thus, the preference information transmitting device 8 may be automatically discovered.

<Operation of Content Recommendation Device 2A>
The content recommendation device 2A according to the second embodiment of the present invention mainly performs preference analysis processing and music recommendation processing. Therefore, each of these processes will be described in detail below.

≪Preference analysis process≫
Details of the preference analysis processing in the preference analysis unit 515 of the content recommendation device 2A according to the second embodiment of the present invention will be described.

  FIG. 9 is a flowchart showing the main processing for one cycle of the preference analysis processing that is periodically operating at a predetermined time interval in the content recommendation device 2A according to the second embodiment of the present invention.

  Here, in the flowchart of the preference analysis process in the content recommendation device 2A according to the second embodiment of the present invention, step S301 is the same in the flowchart of the preference analysis process in the content recommendation device 2 according to the first embodiment of the present invention in FIG. Since the steps are the same as those denoted by reference numerals, description thereof is omitted.

  In step S602, the usage time histogram calculation unit 515b of the preference analysis unit 515 reads the operation history from the operation history storage unit 511.

  Next, the usage time histogram calculation unit 515b of the preference analysis unit 515 calculates a usage time histogram (step S606). Specifically, for the operation history read out in step S602, the number of operation histories whose operation times are included in the time zone is counted for each time zone obtained by dividing 24 hours a day at intervals of, for example, 2 hours. A histogram is obtained in which the number of histories is the frequency of the time period.

  Next, the user identification unit 515c of the preference analysis unit 515 calculates the usage time similarity of each preference information group 401 included in the preference information received in step S301 (step S607). Specifically, each of the usage time histogram 402 included in the preference information group 401 and the usage time histogram calculated in step S606 is represented by a multidimensional vector having a time zone as a dimension and a frequency as a value. The cosine distance of the vector is obtained, and the obtained cosine distance is calculated as the usage time similarity of the preference information group 401.

  Next, the user identification unit 515c of the preference analysis unit 515 selects the preference information group 401 whose usage time similarity obtained in step S607 is equal to or greater than a predetermined threshold (step S608). The preference information received in step S301 may include a preference information group 401 that reflects a preference of a user different from the user of the content recommendation device 2A, but a preference having a usage time histogram similar to the usage time histogram. The information group 401 is considered to have a high possibility of reflecting the same user preference as the user of the content recommendation device 2A, and this step reflects the same user preference as the user of the content recommendation device 2A. Only the preference information group 401 can be selected. The threshold used here is set to an optimum value by experiment.

  Next, the preference information integration unit 515d of the preference analysis unit 515 integrates the plurality of preference information groups 401 selected in step S608 (step S609). Specifically, when there are a plurality of keyword preference levels 403 having the same keyword 406 for the keyword preference levels 403 included in the preference information group 401, the keywords 406 are added by adding the preference levels 407 of the keyword preference levels 403. The new preference degree 407 is assumed. Then, a set of the integrated keyword 406 and preference 407 is associated with each other and stored in the keyword preference storage unit 512.

≪Music recommendation process≫
Next, details of the music recommendation process in the music recommendation unit 506 of the content recommendation device 2A according to the second embodiment of the present invention will be described.

  FIG. 10 is a flowchart showing music recommendation processing in the music recommendation unit 506 of the content recommendation device 2A according to the second embodiment of the present invention.

  When music recommendation is instructed from the device control unit 504, the music recommendation unit 506 starts a music recommendation process and reads the keyword preference degree from the keyword preference degree storage unit 512 (step S701).

  Next, the music recommendation unit 506 rearranges the keywords included in the keyword preference read in step S701 in descending order of the corresponding preference (step S702).

  Next, the music recommendation unit 506 substitutes a predetermined threshold number of keywords for the variable N and substitutes 0 for the variable i (step S703).

  Next, the music recommendation unit 506 determines whether or not the value of the variable i is smaller than the value of the variable N (step S704). If it is determined that the value is smaller than the value of the variable N, the process proceeds to step S705. If it is determined that the value is not smaller than the value of the variable N, that is, the value is not less than the value of the variable N, the process proceeds to step S709.

  If it is determined in step S704 that the value of the variable i is smaller than the value of the variable N, 1 is added to the value of the variable i and substituted for the variable i (step S705).

  Next, the music recommendation unit 506 requests the server device 5 to search for the i-th keyword rearranged in step S702 (step S706).

  Next, the music recommendation unit 506 receives a list of music that is a search result transmitted from the server device 5 (step S707).

  Next, the music recommendation unit 506 integrates the search results received in step S707 (step S708). Specifically, if step S706 and step S707 are executed a plurality of times, there may be duplicate music in the search result music list, but the duplicate music is deleted to create a list of music without duplication To do. Then, the process proceeds to step S704.

  On the other hand, if it is determined in step S704 that the value of the variable i is not smaller than the value of the variable N, that is, the value of the variable N is greater than or equal to the value of the variable N, the music recommendation unit 506 selects the music that is the search result integrated in step S708. The music recommendation screen is created by arranging the lists of the above, and the created music recommendation screen is output to the output control unit 505 to recommend the music as the content to the corresponding user (step S709).

  Therefore, according to the content recommendation device of the second embodiment, the operation history of the device by the user is stored, the user preference information is collected, the user operation history is read, and the usage time histogram is calculated for each user folder. Then, the usage time histogram included in the collected user preference information is compared with the calculated user usage time histogram, and for each user folder, whether the preference information user is the same as the folder user. For each identified user folder, the preference information identified as the same user as the folder user is integrated, and the keyword preference included in the preference information is determined for each integrated user folder. When the recommendation of music is instructed from the user, the keyword preference degree for each user is read, and the read keyword for each user The user requests the server device to search for a keyword included in the preference, receives the music list as a search result transmitted from the server device, and outputs the received music list as a music recommendation screen. It is possible to recommend music that matches the user's preference and interest and that the user has not watched similar music in the past.

  Next, as a third embodiment of the present invention, an example of a preference information transmitting device will be described. In the third embodiment, the content is described as a television program received via a broadcast wave, for example.

<Configuration 1 of preference information transmitting apparatus>
FIG. 11 is a diagram illustrating a configuration example of the preference information transmitting apparatus according to the third embodiment of the present invention. The preference information transmission device 8 shown in FIG. 11 has a function as a television recording device for viewing and recording television programs received via broadcast waves.

  The preference information transmitting apparatus 8 includes an input unit 201, an output unit 202, a broadcast receiving unit 203, a device control unit 804, an output control unit 205, a recording unit 207, a program information storage unit 208, and recording reservation information. A storage unit 209, a video / audio signal recording unit 210, an operation history storage unit 211, a keyword preference storage unit 812, a network connection unit 213, a preference information transmission unit 814, and a preference analysis unit 815 are provided.

  Here, among the configurations provided in the preference information transmission device 8 according to the third embodiment of the present invention, the input unit 201, the output unit 202, the broadcast reception unit 203, the output control unit 205, the recording unit 207, The program information storage unit 208, the recording reservation information storage unit 209, the video / audio signal recording unit 210, the operation history storage unit 211, and the network connection unit 213 are respectively provided in the content recommendation device 2 of the first embodiment. Since it is the same as the structure to which the same code | symbol was attached | subjected, description is abbreviate | omitted.

  Based on the operation signal received from the input unit 201, the device control unit 804 sets a channel received from the broadcast receiving unit 203, stores recording reservation information in the recording reservation information storage unit 209, and records video and audio signals. The reproduction of the video / audio signal recorded in the unit 210 is started, and either the video / audio signal supplied from the broadcast receiving unit 203 or the video / audio signal read out from the video / audio signal recording unit 210 is output to the output control unit 205. Instructs whether to output to the unit 202.

  Here, the recording reservation information is information for executing recording by a timer, and includes information such as a recording channel and recording start / end times. Furthermore, the information of the folder for storing the recorded program is also included. A folder is for a user to organize recorded programs. Particularly when a plurality of users use a TV recording device, a folder corresponding to each user is prepared, and a user selects a favorite program. It records in its own folder.

  In addition, the device control unit 804 causes the operation history storage unit 211 to store information on the recording reservation operation by the user as an operation history. The operation history includes the time when the operation was performed and the recording reservation information, that is, the channel to be recorded, the recording start / end time, the folder information, and the like.

  The keyword preference level storage unit 812 is configured by a storage device such as a hard disk, for example, and stores a plurality of sets of keywords and preference levels and usage time histograms assigned to folders.

  The preference information transmission unit 814 communicates with the content recommendation device 2 connected to the home network 3 via the network connection unit 213. When the content recommendation device 2 requests transmission of preference information, the keyword preference storage unit 812 A plurality of sets of keywords and preference levels assigned to each folder and usage time histograms are read out and transmitted to the content recommendation device 2 as preference information.

  The preference analysis unit 815 executes preference analysis processing at predetermined time intervals such as one day, one hour, one week, etc., and has a keyword extraction unit 815a and a usage time histogram calculation. A unit 815b and a keyword preference calculation unit 815e.

  The keyword extraction unit 815a reads program information from the program information storage unit 208 and extracts keywords from the read program information. Then, the extracted keyword is supplied to the keyword preference calculation unit 815e.

  The usage time histogram calculation unit 815b reads the operation history from the operation history storage unit 211, and calculates a usage time histogram for each folder. Then, the calculated usage time histogram is assigned to a folder and stored in the keyword preference storage unit 812. Further, the read operation history is supplied to the keyword preference calculation unit 815e.

  The keyword preference calculation unit 815e calculates the keyword preference based on the keyword supplied from the keyword extraction unit 815a and the operation history supplied from the usage time histogram calculation unit 815b, and sets a pair of the keyword and the preference level. It is assigned to a folder and stored in the keyword preference storage unit 812.

  The preference information transmitting apparatus 8 according to the third embodiment of the present invention includes a program information storage unit 208, a recording reservation information storage unit 209, a video / audio signal recording unit 210, an operation history storage unit 211, and a keyword preference storage. Although the storage unit 812 is configured as a separate storage device, each storage unit may be partially shared, or all storage units may be shared by a single storage device.

<Operation of Preference Information Transmitting Device 8>
Next, the preference analysis process of the preference information transmission device 8 according to the third embodiment of the present invention will be described in detail.

≪Preference analysis process≫
Details of the preference analysis processing in the preference analysis unit 515 of the preference information transmitting apparatus 8 according to the third embodiment of the present invention will be described.

  FIG. 12 is a flowchart showing the main processing for one cycle of the preference analysis processing that is regularly operating at a predetermined time interval in the preference information transmitting apparatus 8 according to the third embodiment of the present invention. .

  When starting the preference analysis process, the keyword extraction unit 815a of the preference analysis unit 515 reads program information from the program information storage unit 208 (step S901).

  Next, the keyword extraction unit 815a of the preference analysis unit 515 extracts keywords from the program information read in step S901 (step S902). Specifically, “genre” and “performer” included in the program information are extracted as keywords. Further, a sentence such as “title” and “content description” is analyzed using a morphological analysis technique, and morphemes of specific parts of speech such as nouns are extracted as keywords.

  Next, the usage time histogram calculation unit 815b of the preference analysis unit 515 reads the operation history from the operation history storage unit 211 (step S903).

  Next, the number of folders prepared in the preference information transmitting device 8 is substituted into a variable N, and 0 is substituted into a variable i (step S904).

  Next, it is determined whether or not the value of the variable i is smaller than the value of the variable N (step S905). If it is determined that the value is smaller than the value of the variable N, the process proceeds to step S906. If it is determined that the value is not smaller than the value of the variable N, that is, it is equal to or greater than the value of the variable N, the preference analysis process ends.

  If it is determined in step S905 that the value of the variable i is smaller than the value of the variable N, 1 is added to the value of the variable i and substituted for the variable i (step S906).

  Next, the usage time histogram calculation unit 815b of the preference analysis unit 515 calculates a usage time histogram of the folder (step S907). A specific calculation method is the same as that in step S306 of the preference analysis process in the preference analysis unit 215 of the content recommendation device 2. The calculated usage time histogram is assigned to a folder and stored in the keyword preference storage unit 812.

  Next, the keyword preference calculation unit 815e of the preference analysis unit 515 calculates the appearance frequency of the keyword in the operation history (step S908). Specifically, for the operation history in which the folder included in the operation history is the i-th folder among the operation histories read out in step S903, the channels included in the operation history, the recording start / end times, and in step S901 The program corresponding to the operation history is specified by comparing with the read program information. Further, the number of appearances of each keyword extracted in step 902 from the program information of the specified program is counted as one. In this way, the number of appearances of each keyword for all operation histories whose folders included in the operation history are the i-th folder is counted and used as the keyword appearance frequency.

  Next, the keyword preference calculation unit 815e of the preference analysis unit 515 normalizes the keyword appearance frequency obtained in step S908 and calculates the keyword preference (step S909). Specifically, normalization is performed so that the maximum value of the keyword preference degree is 1 by dividing the appearance frequency of each keyword by the maximum value of the appearance frequency. Then, a combination of a keyword and a preference level is assigned to a folder and stored in the keyword preference level storage unit 812. After executing this step, the process proceeds to step S905.

  The above preference analysis processing will be described below in more detail using a specific example.

  The program information for one program read in step S901 is, for example, program information 14 shown in FIG. At this time, keywords such as “world heritage”, “documentary”, “Europe”, “gourmet”, “Jiro Sato”, “Saburo Tanaka” and the like are extracted by extracting keywords in step S902. Such keyword extraction is performed on the program information of all the programs read out in step S901.

  FIG. 13 is a diagram illustrating an example of keyword preference calculation.

  As illustrated in FIG. 13, the keyword column 1501 is an example of keywords extracted in step S902. An appearance frequency column 1502 is an example of the keyword appearance frequency obtained in step S908 for each keyword in the keyword column 1501. Since the maximum value of the appearance frequency is “13” of the keyword “sports”, when the value of each appearance frequency in the appearance frequency column 1502 is divided by 13, the preference level is obtained as shown in the preference level column 1503.

  Then, the preference information transmission unit 814 communicates with the content recommendation device 2 connected to the home network 3 via the network connection unit 213, and when the content recommendation device 2 requests transmission of preference information, the keyword preference degree storage is performed. A plurality of combinations of keywords and preference levels assigned to each folder and usage time histograms are read from the unit 812 and transmitted to the content recommendation device 2 as preference information.

  Therefore, according to the preference information transmitting apparatus 8 of the third embodiment, the operation history of the apparatus by a plurality of users is stored, the program information is stored in advance, the stored program information is read, and the keyword is extracted from the read program information. Extract and read the operation history of multiple users, calculate usage time histograms for multiple folders, assign and store the calculated usage time histograms for multiple folders to multiple user folders and extract A keyword preference level is calculated on the basis of the keyword and the operation history, a pair of the keyword and the preference level is assigned to and stored in a folder of a plurality of users, and the keyword assigned to each folder of the plurality of users A plurality of sets of preference levels and usage time histograms are read, and the content estimation of the first embodiment shown in FIG. Since the data is transmitted to the device 2, the content recommendation device 2A of the second embodiment shown in FIG. 8, the content recommendation device 2B of the fifth embodiment shown in FIG. 17 to be described later, etc., these content recommendation devices 2, 2A, 2B Then, it is possible to recommend a program that matches the user's preference and interest and that the user has not viewed a similar program in the past.

  In the third embodiment, the content is described as a program received via a broadcast wave. However, the present invention is not limited to this, and may be music or the like.

  Next, another example of the preference information transmitting apparatus will be described as a fourth embodiment of the present invention. In the fourth embodiment, as content, for example, music will be described as an example.

<Configuration 2 of preference information transmission device>
FIG. 14 is a diagram illustrating the configuration of the preference information transmitting apparatus 8A according to the fourth embodiment of the present invention. The preference information transmission device 8A shown in FIG. 14 has a function as a music reproduction device that reproduces music received from the server device 5.

  The preference information transmitting apparatus 8A includes an input unit 501, an output unit 202, a music reception unit 1003, a device control unit 1004, an output control unit 505, a music search unit 1006, a music information storage unit 1008, and an audio signal. A recording unit 510, an operation history storage unit 511, a keyword preference storage unit 1012, a network connection unit 213, a preference information transmission unit 1014, and a preference analysis unit 1015 are provided.

  Here, in the configuration provided in the preference information transmitting device 8A of the third embodiment of the present invention, the output unit 202 and the network connection unit 213 are respectively the same reference numerals provided in the content recommendation device 2 of the first embodiment. Since it is the same as the structure to which is attached | subjected, description is abbreviate | omitted.

  Of the configurations provided in the preference information transmitting apparatus 8A according to the third embodiment of the present invention, the input unit 501, the output control unit 505, the audio signal recording unit 510, and the operation history storage unit 511 are each implemented. Since it is the same as the structure to which the same code | symbol provided in content recommendation apparatus 2A of Example 1 was attached | subjected, description is abbreviate | omitted.

  When the music control unit 1004 is instructed to receive music, the music receiving unit 1003 communicates with the server device 5 connected to the external network 4 via the network connection unit 213 via the home network 3, and the server The apparatus 5 is requested to transmit the music, and the audio signal of the music transmitted from the server apparatus 5 is received and recorded in the audio signal recording unit 510. Furthermore, the music information transmitted together with the audio signal of the music from the server device 5 is received and stored in the music information storage unit 1008.

  Here, the music information is information added to the audio signal by the server device side, and includes information such as “title”, “artist name”, “genre”, and “music identification information”.

  The device control unit 1004 instructs the music receiving unit 503 to receive music based on the operation signal received from the input unit 501 and starts playing the audio signal recorded in the audio signal recording unit 510. The output control unit 505 is instructed to output the audio signal read from the audio signal recording unit 510 to the output unit 202.

  In addition, the device control unit 1004 stores information on the music reproduction operation by the user in the operation history storage unit 511 as an operation history.

  Here, the operation history includes time when the operation is performed, music identification information for identifying the music, and the like.

  In addition, the device control unit 1004 instructs the music search unit 1006 to search for the music together with the search condition based on the operation signal received from the input unit 501, and the music search supplied from the music search unit 1006 to the output control unit 505. Instructs the output of the result screen.

  Here, the search condition is a keyword such as a song title or artist name.

  The music search unit 1006 communicates with the server device 5 connected to the outside network 4 via the network connection unit 213 via the home network 3 when a search is instructed together with the search condition from the device control unit 1004. The server device 5 is requested to perform a search based on a search condition keyword, a music list that is a search result transmitted from the server device 5 is received, and the received music list is supplied to the output control unit 505 as a music search result screen. is there.

  The music information storage unit 1008 is configured by a storage device such as a hard disk, for example, and stores music information supplied from the music receiving unit 1003.

  The keyword preference storage unit 1012 is constituted by a storage device such as a hard disk, for example, and stores a set of keywords and preference and a usage time histogram.

  The preference information transmission unit 1014 communicates with the content recommendation device 2 connected to the home network 3 via the network connection unit 213, and when the content recommendation device 2 requests transmission of preference information, the keyword preference storage unit 1012. From the keyword and the preference level and the usage time histogram are read out and transmitted to the content recommendation device 2 as preference information.

  The preference analysis unit 1015 performs preference analysis processing at predetermined time intervals such as one day, one hour, one week, and the like. For example, the preference analysis unit 1015 includes a usage time histogram calculation unit 1015b and an artist preference. Degree calculator 1015e.

  The usage time histogram calculation unit 1015b reads an operation history from the operation history storage unit 511, calculates a usage time histogram, and stores the calculated usage time histogram in the keyword preference storage unit 1012. Further, the read operation history is supplied to the artist preference level calculation unit 1015e.

  The artist preference level calculation unit 1015e calculates the preference level of the artist based on the operation history supplied from the usage time histogram calculation unit 1015b, and sets the keyword and preference level as a keyword in the keyword preference level storage unit 1012 using the artist name as a keyword. It is something to remember.

  The preference information transmitting apparatus 8A according to the fourth embodiment of the present invention includes a music information storage unit 1008, an audio signal recording unit 510, an operation history storage unit 511, and a keyword preference level storage unit 1012 as separate storage devices. Although configured, each storage unit may be partially shared, or all the storage units may be shared by one storage device.

<Operation of Preference Information Transmitting Device 8A>
Next, the preference analysis process of the preference information transmitting apparatus 8A according to the fourth embodiment of the present invention will be described in detail.

≪Preference analysis process≫
Details of the preference analysis processing in the preference analysis unit 1015 of the preference information transmission device 8A according to the fourth embodiment of the present invention will be described.

  FIG. 15 is a flowchart illustrating main processing for one cycle of the preference analysis processing that is periodically operated at a predetermined time interval in the preference information transmission device 8A according to the third embodiment of the present invention. .

  The artist preference level calculation unit 1015e of the preference analysis unit 1015 reads music information from the music information storage unit 1008 when the preference analysis processing is started (step S1101).

  Next, the usage time histogram calculation unit 1015b of the preference analysis unit 1015 reads the operation history from the operation history storage unit 511 (step S1102).

  Next, the usage time histogram calculation unit 1015b of the preference analysis unit 1015 calculates a usage time histogram (step S1106). A specific calculation method is the same as that in step S606 of the preference analysis process in the preference analysis unit 515 of the content recommendation device 2A. The calculated usage time histogram is stored in the keyword preference storage unit 1012.

  Next, the artist preference level calculation unit 1015e of the preference analysis unit 1015 calculates the appearance frequency of the artist in the operation history (step S1107). Specifically, the artist preference level calculation unit 1015e corresponds to the operation history by comparing the music identification information included in the operation history with the music information read in step S1101 for the operation history read in step S1102. Identify the song. Furthermore, the number of appearances of the artist included in the music information of the specified music is counted as one. In this way, the number of appearances of each artist for all operation histories is counted as the appearance frequency of the artist.

  Next, the artist preference level calculation unit 1015e of the preference analysis unit 1015 normalizes the appearance frequency of the artist obtained in step S1107, and calculates the artist preference level (step S1108). Specifically, normalization is performed so that the maximum value of artist preference is 1 by dividing the appearance frequency of each artist by the maximum value of the appearance frequency. Then, a set of a keyword and a preference level is stored in the keyword preference level storage unit 1012 using the artist name as a keyword.

  Then, the preference information transmission unit 1014 communicates with the content recommendation device 2 connected to the home network 3 via the network connection unit 213, and when the content recommendation device 2 requests transmission of preference information, the keyword preference degree storage is performed. The combination of the artist name and the preference level as a keyword and the usage time histogram are read from the unit 1012 and transmitted to the content recommendation device 2 as preference information.

  Therefore, according to the preference information transmitting apparatus of the fourth embodiment, the operation history of the apparatus by a plurality of users is stored, the music information is stored in advance, the operation histories of the plurality of users are read, and the plurality of users are Calculate usage time histograms, store the calculated usage time histograms for multiple folders in multiple user folders, calculate the preference for each artist based on the operation history of multiple users, and calculate The preference level for each artist is stored in association with the preference level using the artist name as a keyword and assigned to a folder of a plurality of users as a set, and the combination of the keyword and the preference level assigned to each folder of the plurality of users And the usage time histogram, and the content recommendation device 2 of the first embodiment shown in FIG. The content recommendation device 2A of the second embodiment shown in FIG. 17 and the content recommendation device 2B of the fifth embodiment shown in FIG. 17 to be described later are transmitted to the content recommendation device 2, 2A, 2B. It is possible to recommend music that fits the user's interest and that the user has not viewed similar music in the past.

  In the fourth embodiment, the content is described as music, but the present invention is not limited to this, and may be a program or the like.

  Next, a content recommendation device 2B will be described as a fifth embodiment of the present invention.

  FIG. 16 is a diagram illustrating a connection example between the content recommendation device 2B and the server device 5 according to the fifth embodiment of this invention.

  The content recommendation device 2B according to the fifth embodiment is connected to each other via the home network 3, and the home network 3 is connected to the outside network 4 and is connected to the server device 5 via the outside network 4. In FIG. 16, the connection destination of the content recommendation device 2B of the fifth embodiment connected to each other via the home network 3 is the content recommendation device 2 of the first embodiment, the content recommendation device 2A of the second embodiment, Of course, the preference information transmission device 8 according to the third embodiment and the preference information transmission device 8A according to the fourth embodiment may be used.

  In FIG. 16, the in-home network 3, the out-of-home network 4, and the server device 5 are the same as those in the first embodiment shown in FIG.

  The content recommendation device 2B according to the fifth embodiment has a function as a content viewing device that receives and views content such as a television program transmitted via a broadcast wave and a moving image or music transmitted from the server device 5. Furthermore, the user's preference is estimated based on the preference information received from the other content recommendation device 2B via the home network 3, and the content that matches the user's preference is recommended. Furthermore, preference information is generated based on a history of operations performed on the content recommendation device 2B by the user, and preference information is transmitted in response to a request from another content recommendation device 2B connected via the home network 3.

  In FIG. 16, the server device 5 may be omitted. In this case, the content recommendation device 2B views content such as television programs transmitted via broadcast waves, and content such as moving images and music recorded on physical media such as CDs and DVDs.

<Configuration of content recommendation device>
FIG. 17 is a diagram illustrating a configuration example of the content recommendation device 2B according to the fifth embodiment of the present invention. The content recommendation device 2B shown in FIG. 17 has a function as a television recording device for viewing and recording a television program received via a broadcast wave.

  The content recommendation device 2B includes an input unit 201, an output unit 202, a broadcast reception unit 203, a device control unit 204, an output control unit 205, a program recommendation unit 206, a recording unit 207, and a program information storage unit 208. A recording reservation information storage unit 209, a video / audio signal recording unit 210, an operation history storage unit 211, a program preference storage unit 212, a network connection unit 213, a preference information reception unit 214, and a keyword preference storage. A unit 812, a preference information transmission unit 814, and a preference analysis unit 1715.

  Here, among the configurations provided in the content recommendation device 2B according to the fifth embodiment of the present invention, the input unit 201, the output unit 202, the broadcast receiving unit 203, the device control unit 204, the output control unit 205, Program recommendation unit 206, recording unit 207, program information storage unit 208, recording reservation information storage unit 209, video / audio signal recording unit 210, operation history storage unit 211, program preference storage unit 212, network The connection unit 213 and the preference information receiving unit 214 are the same as the configurations with the same reference numerals provided in the content recommendation device 2 of the first embodiment, and thus the description thereof is omitted. Of the configurations provided in the content recommendation device 2B according to the fifth embodiment of the present invention, the keyword preference storage unit 812 and the preference information transmission unit 814 are provided in the preference information transmission device 8 according to the first embodiment. Since the configuration is the same as that denoted by the same reference numeral, the description thereof is omitted.

  The preference analysis unit 1715 performs preference analysis processing at predetermined time intervals such as one day, one hour, one week, etc., and has a keyword extraction unit 815a and usage time histogram calculation. A unit 1715b, a keyword preference calculation unit 815e, a preference information collection unit 215a, a user identification unit 215c, a preference information integration unit 1715d, and a program preference calculation unit 215e.

  Here, among the configurations provided in the preference analysis unit 1715 of the content recommendation device 2B according to the fifth embodiment of the present invention, the preference information collection unit 215a, the user identification unit 215c, and the program preference level calculation unit 215e are respectively Since it is the same as the structure to which the same code | symbol provided in the preference analysis part 215 of the content recommendation apparatus 2 of Example 1 was attached | subjected, description is abbreviate | omitted. Of the configurations provided in the preference analysis unit 1715 of the content recommendation device 2B according to the fifth embodiment of the present invention, the keyword extraction unit 815a and the keyword preference degree calculation unit 815e are respectively preference information according to the third embodiment. Since it is the same as the structure to which the same code | symbol provided in the preference analysis part 815 of the transmitter 8 was attached | subjected, description is abbreviate | omitted.

  The usage time histogram calculation unit 1715b reads the operation history from the operation history storage unit 211 and calculates a usage time histogram for each folder. Then, the calculated usage time histogram is assigned to a folder and stored in the keyword preference storage unit 812, and the calculated usage time histogram is supplied to the user identification unit 215c. Further, the read operation history is supplied to the keyword preference calculation unit 815e.

  The preference information integration unit 1715d reads a keyword / preference degree set assigned to a folder from the keyword preference storage unit 812 and, for each folder, the keyword and preference assigned to the folder read from the keyword preference storage unit 812. And the preference information identified as being the same user as the user of the folder are integrated, and the integrated preference information is supplied to the program preference level calculation unit 215e.

≪Preference analysis process≫
Next, details of the preference analysis processing in the preference analysis unit 1715 of the content recommendation device 2B according to the fifth embodiment of the present invention will be described.

  FIG. 18 is a flowchart showing the main processing for one cycle of the preference analysis processing that is periodically operating at a predetermined time interval in the content recommendation device 2B according to the fifth embodiment of the present invention.

  Here, in the flowchart of the preference analysis process in the content recommendation device 2B according to the fifth embodiment of the present invention, step S307, step S308, step S310, and step S311 are the first embodiment of the present invention shown in FIG. Since it is the same as the step which attached | subjected the same code | symbol in the flowchart of the preference analysis process in a certain content recommendation apparatus 2, description is abbreviate | omitted. Of the flowchart of the preference analysis process in the content recommendation device 2B according to the fifth embodiment of the present invention, step S901, step S902, step S903, step S904, step S905, step S906, and step S908 Step S909 is the same as the step denoted by the same reference numeral in the flowchart of the preference analysis process in the preference information transmitting apparatus 8 of the third embodiment shown in FIG.

  In step S1801, the preference information collection unit 215a of the preference analysis unit 1715 receives preference information from another content recommendation device 2 when the preference analysis process is started. Specifically, the preference information collection unit 215a instructs the preference information reception unit 214 to receive preference information and receives the preference information supplied from the preference information reception unit 214.

  In step S1807, the usage time histogram calculation unit 1715b of the preference analysis unit 1715 calculates a usage time histogram of the folder. A specific calculation method is the same as that in step S306 of the preference analysis process in the preference analysis unit 215 of the content recommendation device 2. The calculated usage time histogram is assigned to a folder and stored in the keyword preference storage unit 812, and the calculated usage time histogram is supplied to the user identification unit 215c.

  In step S1809, the preference information integration unit 1715d of the preference analysis unit 1715 integrates the keyword preference calculated in step S909 and the plurality of preference information groups 401 selected in step S308. Specifically, there are a plurality of keyword preference levels 403 having the same keyword 406 for the keyword preference levels calculated in step S909 and the keyword preference levels 403 included in the plurality of preference information groups 401 selected in step S308. In this case, the preference degree 407 of the keyword preference degree 403 is added to obtain a new preference degree 407 of the keyword 406.

  In the content recommendation device 2B according to the fifth embodiment, when the preference analysis processing in the preference analysis unit 1715 ends, and when the recommendation of the program is instructed together with the folder information from the device control unit 204, the program recommendation unit 206 displays the program information The program information is read from the storage unit 208, the program preference level corresponding to the information of the folder instructed from the device control unit 204 is read from the program preference level storage unit 212, and the program preference level is higher among the programs to be broadcast in the future. A predetermined number of programs are selected, a list of program information of the selected programs is supplied to the output unit 202 as a program recommendation screen, and programs corresponding to contents are recommended to corresponding users.

  Further, in the content recommendation device 2B of the fifth embodiment, the preference analysis processing in the preference analysis unit 1715 is finished, communicates with the content recommendation device 2 connected to the home network 3 via the network connection unit 213, and other content recommendation When transmission of preference information is requested from the device, the preference information transmission unit 814 reads a plurality of pairs of keywords and preference levels assigned to each folder and usage time histograms from the keyword preference level storage unit 812, It transmits to the said other content recommendation apparatus as preference information.

  Therefore, according to the content recommendation device 2B of the fifth embodiment, the operation history of the device by a plurality of users is stored, the content information is stored in advance, the content information is read, the keyword is extracted from the read content information, Read operation history of multiple users, calculate usage time histograms for multiple folders, assign and store calculated usage time histograms for multiple folders to multiple user folders, extracted keywords and operation history Based on the above, the preference degree of the keyword is calculated, a set of the keyword and the preference degree is assigned to and stored in a folder of a plurality of users, and the preference information of the plurality of users is collected, The usage time histogram included in the preference information is compared with the usage time histograms of multiple users. For each folder, it is identified whether the user of the preference information is the same as the user of the folder, and the preference information that is identified as the same user as the user of each folder is integrated for each folder of the plurality of identified users. Based on the preference information integrated for each folder, the content preference level of the content scheduled for the future broadcast is calculated for each folder, the content preference level of the content scheduled for the future broadcast calculated for each folder is stored, Since the content preference level of the content is read out, a predetermined number of contents are selected from the contents that are broadcast in the future, and the list of content information of the selected content is recommended as the content recommendation screen. Content that matches the user's preferences and interests It is possible to recommend content that is not view, to other content recommendation device can transmit the preference information.

  In the first, second, and fifth embodiments, it is assumed that the content recommendation devices 2, 2A, and 2B have functions as a television recording device for viewing / recording a television program or a music reproduction device for reproducing a song, respectively. Although assumed, it is not limited to these forms, You may comprise the apparatus which views a content with another form as content recommendation apparatus 2,2A, 2B. Also, the content is not limited to a broadcast program or music, but may of course include text.

  In the third and fourth embodiments, the preference information transmitting devices 8 and 8A are assumed to have a function as a television recording device for viewing and recording television programs or a music reproducing device for reproducing music. However, the present invention is not limited to these forms, and other forms of apparatuses capable of generating user preference information may be configured as the preference information transmission apparatuses 8 and 8A.

  In the first, second, and fifth embodiments, the content recommendation devices 2, 2A, and 2B can be realized by executing a content recommendation program installed in a computer. The content recommendation program may be read from a recording medium storing the content recommendation program and executed by the CPU to constitute a content recommendation device, or may be transmitted via a communication network. The content recommendation devices 2, 2A, and 2B may be configured by being installed and executed by the CPU.

  In the third and fourth embodiments, the preference information transmitting devices 8 and 8A can also be realized by executing a preference information transmitting program installed in a computer. For example, the preference information transmission program may be read from a recording medium in which the preference information transmission program is stored and executed by the CPU to configure the preference information transmission device, or via a communication network. The preference information transmitting devices 8 and 8A may be configured by being transmitted, installed, and executed by the CPU.

  In the first to fifth embodiments, a folder is provided for each user, and the user is identified by the folder. However, the present invention is not limited to this, and the user name, user ID, and the like are identified. Of course, the user may be identified based on the information that can be obtained.

2, 2A, 2B ... Content recommendation device 5 ... Server device 8, 8A ... Preference information transmission device 201 ... Input unit 202 ... Output unit 203 ... Broadcast reception unit 204 ... Device control unit 205 ... Output control unit 206 ... Program recommendation unit 207 ... Recording unit 208 ... Program information storage unit 209 ... Recording reservation information storage unit 210 ... Video / audio signal recording unit 211 ... Operation history storage unit 212 ... Program preference storage unit 213 ... Network connection unit 214 ... Preference information reception unit 215 ... Preferences Analysis unit 215a ... Preference information collection unit 215b ... Usage time histogram calculation unit 215c ... User identification unit 215d ... Preference information integration unit 215e ... Program preference level calculation unit 501 ... Input unit 503 ... Music reception unit 504 ... Device control unit 505 ... Output Control unit 506 ... Music recommendation unit 510 ... Audio signal recording unit 511 ... Operation history storage unit 512 ... Key Word preference level storage unit 515 ... Preference analysis unit 515b ... Usage time histogram calculation unit 515c ... User identification unit 515d ... Preference information integration unit 804 ... Device control unit 812 ... Keyword preference level storage unit 814 ... Preference information transmission unit 815 ... Preference analysis Unit 815a ... keyword extraction unit 815b ... use time histogram calculation unit 815e ... keyword preference calculation unit 1003 ... music reception unit 1004 ... equipment control unit 1006 ... music search unit 1008 ... music information storage unit 1012 ... keyword preference storage unit 1014 ... Preference information transmission unit 1015, 1715 ... Preference analysis unit 1015b, 1715b ... Usage time histogram calculation unit 1015e ... Artist preference level calculation unit 1715d ... Preference information integration unit 5204 ... Device control unit

Claims (6)

Operation history storage means for storing operation history of the content recommendation device by one or more users;
Preference information collecting means for collecting preference information of the one or more users;
A usage time histogram calculation unit that reads the operation history of the one or more users from the operation history storage unit, and calculates a usage time histogram for each of the one or more users based on the read operation history;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection means with the usage time histogram of the one or more users calculated by the usage time histogram calculation means; User identification means for identifying whether the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation means;
Preference information integration means for integrating the preference information for each user identified as the same when the user is identified by the user identification means;
Content that calculates the content preference level of content scheduled to be broadcast in the future for each user identified as the same based on the preference information integrated for each user identified as the same by the preference information integration means A preference calculation means;
Content preference storage means for storing the content preference of content scheduled to be broadcast in the future calculated for each user identified as the same;
When content recommendation is requested by the user's operation, the content preference level of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage unit, and the content preference level is determined based on the content preference level. Content recommendation means for recommending,
A content recommendation device comprising: Operation history storage means for storing operation history of the content recommendation device by one or more users;
Preference information collecting means for collecting preference information of the one or more users;
A usage time histogram calculation unit that reads the operation history of the one or more users from the operation history storage unit, and calculates a usage time histogram for each of the one or more users based on the read operation history;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection means with the usage time histogram of the one or more users calculated by the usage time histogram calculation means; User identification means for identifying whether the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation means;
Preference information integration means for integrating the preference information for each user identified as the same when the user is identified by the user identification means;
Keyword preference storage means for storing the keyword preference included in the preference information for each user identified as the same integrated by the preference information integration means;
When content recommendation is requested by the user's operation, the keyword preference degree corresponding to the operated user is read from the keyword preference degree storage unit, and a search by a keyword included in the read keyword preference degree is performed on the server device. Content recommendation means for requesting and receiving a content list as a search result transmitted from the server device, and outputting the received content list as a content recommendation screen;
A content recommendation device comprising: Operation history storage means for storing operation history of the content recommendation device by one or more users;
Content storage means for storing content information;
Keyword extraction means for reading the content information from the content storage means and extracting keywords from the read content information;
A usage time histogram calculation unit that reads the operation history of the one or more users from the operation history storage unit, and calculates a usage time histogram for each of the one or more users based on the read operation history;
Keyword preference degree calculation means for calculating a preference degree of the keyword based on the keyword extracted by the keyword extraction means and the operation history read from the operation history storage means;
The keyword and the keyword preference level calculation unit calculated by the keyword preference level calculation unit and the usage time histogram for each of the one or more users calculated by the usage time histogram calculation unit are represented by the one or more users. Keyword preference storage means for allocating and storing each;
Preference information collecting means for collecting preference information of the one or more users;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection means with the usage time histogram of the one or more users calculated by the usage time histogram calculation means; User identification means for identifying whether the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation means;
For each user identified as identical by the user identification means when the user identification means identifies the same user as the keyword preference degree storage means and reads out the keyword and the keyword preference degree set. In addition, preference information integration means for integrating the set of the read keyword and the preference degree of the keyword into the preference information;
Content that calculates the content preference level of content scheduled to be broadcast in the future for each user identified as the same based on the preference information integrated for each user identified as the same by the preference information integration means A preference calculation means;
Content preference storage means for storing the content preference of content scheduled to be broadcast in the future calculated for each user identified as the same;
When content recommendation is requested by the user's operation, the content preference level of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage unit, and the content preference level is determined based on the content preference level. Content recommendation means for recommending,
A content recommendation device comprising: An operation history storage step of storing an operation history of the content recommendation device by one or more users in the operation history storage means;
A preference information collecting step for collecting preference information of the one or more users;
A usage time histogram calculation step of reading the operation history of the one or more users from the operation history storage unit, and calculating a usage time histogram for each of the one or more users based on the read operation history;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection step with the usage time histogram of the one or more users calculated by the usage time histogram calculation step; A user identifying step for identifying whether or not the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation step;
When the user identification step identifies the same user, the preference information integration step of integrating the preference information for each user identified as the same;
Content that calculates the content preference level of content scheduled to be broadcast in the future for each user identified as the same based on the preference information integrated for each user identified as the same in the preference information integration step A preference calculation step;
A content preference level storage step of storing in the content preference level storage means the content preference level of content scheduled to be broadcast in the future calculated for each user identified as the same;
When content recommendation is requested by the user's operation, the content preference level of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage unit, and the content preference level is determined based on the content preference level. A content recommendation step for recommending,
Content recommendation program to make computer execute. An operation history storage step of storing an operation history of the content recommendation device by one or more users in the operation history storage means;
A preference information collecting step for collecting preference information of the one or more users;
A usage time histogram calculation step of reading the operation history of the one or more users from the operation history storage unit, and calculating a usage time histogram for each of the one or more users based on the read operation history;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection step with the usage time histogram of the one or more users calculated by the usage time histogram calculation step; A user identifying step for identifying whether or not the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation step;
When the user identification step identifies the same user, the preference information integration step of integrating the preference information for each user identified as the same;
A keyword preference storage step for storing a keyword preference level included in the preference information in a keyword preference level storage unit for each user identified as the same integrated in the preference information integration step;
When content recommendation is requested by the user's operation, the keyword preference degree corresponding to the operated user is read from the keyword preference degree storage unit, and a search by a keyword included in the read keyword preference degree is performed on the server device. A content recommendation step of requesting and receiving a content list which is a search result transmitted from the server device, and outputting the received content list as a content recommendation screen;
Content recommendation program to make computer execute. An operation history storage step of storing an operation history of the content recommendation device by one or more users in the operation history storage means;
A content storage step of storing content information in the content storage means;
A keyword extraction step of reading the content information from the content storage means and extracting a keyword from the read content information;
A usage time histogram calculation step of reading the operation history of the one or more users from the operation history storage unit, and calculating a usage time histogram for each of the one or more users based on the read operation history;
A keyword preference calculation step for calculating a preference level of the keyword based on the keyword extracted by the keyword extraction step and the operation history read from the operation history storage unit;
The keyword and the keyword preference level calculated in the keyword preference level calculation step and the usage time histogram for each of the one or more users calculated in the usage time histogram calculation step are the one or more users. A keyword preference storage step that is assigned and stored in the keyword preference storage means;
A preference information collecting step for collecting preference information of the one or more users;
Comparing the usage time histogram included in the preference information of the one or more users collected by the preference information collection step with the usage time histogram of the one or more users calculated by the usage time histogram calculation step; A user identifying step for identifying whether or not the user of the usage time histogram included in the preference information is the same as the user of the usage time histogram calculated by the usage time histogram calculation step;
For each user identified as identical by the user identification step, when the user identification step identifies the same user, the keyword preference storage unit reads out the keyword and the keyword preference set. A preference information integration step of integrating the set of the read keyword and the preference level of the keyword into the preference information;
Content that calculates the content preference level of content scheduled to be broadcast in the future for each user identified as the same based on the preference information integrated for each user identified as the same in the preference information integration step A preference calculation step;
A content preference level storage step of storing in the content preference level storage means the content preference level of content scheduled to be broadcast in the future calculated for each user identified as the same;
When content recommendation is requested by the user's operation, the content preference level of the content scheduled to be broadcast in the future corresponding to the operated user is read from the content information storage unit, and the content preference level is determined based on the content preference level. A content recommendation step for recommending,
Content recommendation program to make computer execute.

Images