JP6937235B2 - Information processing equipment, information processing methods, and programs - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program.

In recent years, when a user accesses a website via the Internet, the content to be posted on the website is sorted out for each user. For example, by predicting the click rate for each content based on the user's past access history information and posting the content in descending order of the predicted value, a technique for recommending the content suitable for the user's interest and interest is provided. (See, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2017-49941

In the above-mentioned content distribution mechanism, the work of uploading new content is performed by a website editor or the like. On the other hand, the content actually delivered to the user is determined based on the predicted value such as the click rate, and the access record is reflected in the prediction, so the posting position of the newly submitted content is set to the lower position. In some cases, it was not delivered to the user at all. For this reason, the content submission work may be wasted.

In addition, even if the content is related to a topic that many users are expected to be interested in, if a large number of content related to this topic has already been submitted, it will be newly created in relation to these many submitted contents. In some cases, it was difficult to select the content submitted to the site as a distribution target.

The present invention has been made in consideration of such circumstances, and one of the objects of the present invention is to provide an information processing device, an information processing method, and a program capable of appropriately assisting the content registration work. do.

One aspect of the present invention relates to a first generation unit that generates first feature information corresponding to a first content that is a candidate to be distributed via a network, and at least one second content distributed via a network. The second generation unit that generates the second feature information corresponding to the user based on the access history of the user, and the user's interest in the first content based on the first feature information and the second feature information. It is an information processing device including a calculation unit that calculates an index value that predicts the degree.

According to one aspect of the present invention, it is possible to appropriately assist the content registration work.

It is a figure which shows an example of the use environment of the drafting apparatus 30 which concerns on embodiment. It is a figure which shows an example of the content of the user information 22 which concerns on embodiment. It is a figure which shows an example of the content of the distribution article information 24 which concerns on embodiment. It is a figure which shows an example of the content of the action log 26 which concerns on embodiment. It is a functional block diagram which shows the functional structure of the drafting apparatus 30 which concerns on embodiment. It is a figure which conceptually shows the content of the process of the article vector generation unit 37 which concerns on embodiment. It is a figure which conceptually shows the content of the process of the user vector generation unit 39 which concerns on embodiment. It is a flowchart which shows an example of the processing flow of the submission apparatus 30 which concerns on embodiment. It is a figure which shows an example of the submission page which concerns on embodiment. It is a figure which shows another example of the trafficking page which concerns on embodiment.

Hereinafter, embodiments of the information processing apparatus, information processing method, and program of the present invention will be described with reference to the drawings.

[Overview]
The information processing device of the embodiment accepts registration of candidate contents provided to the user. This information processing device is operated by an operator or the like who performs content registration work. The registered contents include articles such as news of various genres, listing information on auction sites, and the like. An article is an electronic article provided via a network such as the Internet, and includes text information.

The above article may be provided as part of a web page played by a browser or as part of an app page displayed by an application program. These web pages and application pages are pages constituting a search site, a news site, a shopping site, an auction site, a weather forecast site, an SNS (Social Networking Service) site, a game site, and the like.

In the following embodiment, an example in which the information processing device is a trafficking device that accepts trafficking of articles will be described. Uploading is the process of rewriting an article created in an external medium into a short summary and registering it in the uploading device, the process of registering an article created in an external medium in the uploading device as it is, and a new operator. It includes the process of registering the created article in the submission device.

[overall structure]
FIG. 1 is a diagram showing an example of a usage environment of the submission device 30. The submission device 30 is connected to the network NW. In addition to the submission device 30, for example, at least one terminal device 10 and at least one service providing device 20 are connected to the network NW. The network NW includes, for example, a part or all of the Internet, WAN (Wide Area Network), LAN (Local Area Network), provider equipment, wireless base station, dedicated line, and the like.

The terminal device 10 is operated by a user who uses a service such as a web page provided by the service providing device 20. The terminal device 10 is, for example, a personal computer, a mobile phone such as a smartphone, a tablet terminal, a PDA (Personal Digital Assistant), or the like.

When the terminal device 10 receives a predetermined operation from the user, the terminal device 10 displays the web page provided by the service providing device 20 by the browser installed in advance. This web page contains at least one article. Further, when the terminal device 10 receives a predetermined operation from the user, the terminal device 10 communicates with the service providing device 20 via the application installed in advance, and acquires the content to be displayed or reproduced on the application. The content is, for example, video data, image data, audio data, text data, or the like, and includes at least one article. As a result, the terminal device 10 is provided with a service similar to the service provided by the various web pages described above via the application.

The service providing device 20 may be a web server device that provides a web page such as the search site described above on the Internet, or communicates with the terminal device 10 in which the application is started to exchange various information. It may be an application server device.

The service provided by the service providing device 20 is provided as a customized service for more individual users by, for example, entering a user ID and a password to log in. When the service providing device 20 is a web server, the user can receive the web page without logging in, but in that case, a general-purpose web page that has not been customized for the user is provided.

When the user logs in, the service providing device 20 posts, for example, an article recommended to the user specified by the user ID on a predetermined screen (for example, the top page of a website or an application page) and is a terminal device. Send to 10. The service providing device 20 determines an article recommended to the user based on, for example, an article vector and a user vector described later. As a result, articles that are close to the user's interests and interests are provided to the user, and the attractiveness of the web page or application page can be maintained and improved.

The service providing device 20 stores the user information 22, the delivered article information 24, and the action log 26 in a storage device (not shown) inside or outside the device.

FIG. 2 is a diagram showing an example of the contents of the user information 22. The user information 22 is, for example, information in which a password, age, gender, birthday, occupation, and other demographic information are associated with a user ID, which is user identification information. The user ID is a login ID used when logging in to the website provided by the service providing device 20. The user ID may be a cookie (HTTP cookie) managed for each web browser provided in the terminal device 10, an IP address of the terminal device 10, or the like.

FIG. 3 is a diagram showing an example of the contents of the delivered article information 24. The delivered article information 24 is, for example, information in which article data is associated with an article ID, which is article identification information. The article data is, for example, text data, image data, a combination thereof, or reference information such as a URL (Uniform Resource Locator).

FIG. 4 is a diagram showing an example of the contents of the action log 26. The action log 26 is, for example, information (access history) in which the browsing page ID, which is the identification information of the browsed page, the article ID of the article included in the browsed page, the browsing time, and the like are associated with each other for each user ID. be.

The submission device 30 is a device that accepts submission of article candidates (hereinafter, referred to as “submission candidate articles”) to be delivered to users. The trafficking device 30 is operated by, for example, an operator who is an editor who uploads articles. The submission device 30 transmits the article data of the article submitted by the editor and the newly assigned article ID to the service providing device 20. As a result, the information about the submitted article is stored as the distribution article information 24 in the storage device in the service providing device 20.

The trafficking device 30 provides a tool that is an interface when the trafficking work is performed by the editor. This interface provides editors with information on how many users would be recommended if this article was actually submitted when they entered a candidate article. For example, this interface provides editors with information on the number of users whose submission candidate articles are displayed at the top of the web page as a result of the articles contained in the web page being customized for each user.

[Configuration of trafficking device]
Hereinafter, the configuration of the submission device 30 will be described. FIG. 5 is a functional block diagram showing a functional configuration of the drafting device 30. The submission device 30, for example, has a communication unit 31, a reception unit 33, a display unit 35, an article vector generation unit 37 (first generation unit), a user vector generation unit 39 (second generation unit), and an order. It includes a determination unit 41 (determination unit), a calculation unit 43, a screen control unit 45 (third generation unit), and a storage unit 47.

The communication unit 31 includes, for example, a communication interface such as a NIC. The communication unit 31 communicates with the service providing device 20 and the like via the network NW.

The reception unit 33 receives an operation input by the editor. The reception unit 33 accepts the input of the information of the submission candidate article (first content) submitted by the editor. The reception unit 33 is, for example, a touch panel, operation buttons, a keyboard, a mouse, or the like.

The display unit 35 displays an operation screen or the like that serves as an interface when the article is submitted. The display unit 35 is, for example, a display or the like. When the display unit 35 is a touch panel compatible display, the display unit 35 may have the above-mentioned functions of the reception unit 33.

The article vector generation unit 37 generates, for example, an article vector (first feature information) corresponding to an article. The article vector is vector data generated by, for example, extracting words included in an article by morphological analysis or the like, and then inputting a word vector distributed and expressed by a method such as word2vec into the first model.

FIG. 6 is a diagram conceptually showing the contents of the processing of the article vector generation unit 37. The content (text) of the article is input to the article vector generation unit 37. The input article includes, for example, a submitted article (second content) acquired from the distribution article information 24 and a submission candidate article input via the reception unit 33.

The article vector generation unit 37 first performs morphological analysis or the like on the content of the article and divides it into words (morphemes). Next, the article vector generation unit 37 applies a method such as word2vec to each word to acquire a distributed expression word vector. One or both of the morphological analysis and word2vec may be realized by requesting the processing to an external device instead of the article vector generation unit 37 performing the processing by itself.

Then, the article vector generation unit 37 derives the article vector by inputting one or more word vectors obtained from the article into the first model. The first model is, for example, a model generated by machine learning. The first model is generated by finding the parameters of the activation function of the hidden layer so that the relationship between the known word vector and the article vector can be reproduced by, for example, a method such as deep learning. Notwithstanding the above definition, the first model may include the identification of word2vec, that is, what kind of corpus is given to word2vec.

The article vector generation unit 37 stores the generated article vector as the article vector information 51 in the storage unit 47 together with the article ID.

The user vector generation unit 39 generates a user vector (second feature information) based on, for example, the history of articles viewed by the user. It can be said that the user vector indicates the object of interest of the user. The user vector is generated using, for example, the above article vector. For example, the user vector is generated by inputting a set of article vectors generated for each of the articles viewed by the user into the second model. The user vector is also vector data represented by a distributed representation.

FIG. 7 is a diagram conceptually showing the contents of the processing of the user vector generation unit 39. The user ID and the article ID of the article viewed by the user indicated by the user ID are input to the user vector generation unit 39. The article ID of the article viewed by the user is acquired from, for example, the action log 26.

First, the user vector generation unit 39 acquires an article vector corresponding to the article ID of the article viewed by the user from the article vector information 51. Next, the user vector generation unit 39 generates a user vector by inputting one or more acquired article vectors into the second model. The second model is, for example, a model generated by machine learning. The second model is generated by finding the parameters of the activation function of the hidden layer so that the relationship between the known article vector and the user vector can be reproduced by, for example, a method such as deep learning.

The user vector generation unit 39 stores the generated user vector in the user vector information 53 together with the user ID. The user vector may be created by reflecting elements other than the article vector of the browsed article. The elements other than the article vector are, for example, a part or all of the query entered by the user, the browsing history of the web page having no article, and the attributes of the user (age, gender, occupation, etc.).

The ranking determination unit 41 determines the posting order of the articles included in the web page for each of the users included in the sample user information 49 when the submitted articles and the submission candidate articles are targeted for distribution. That is, the ranking determination unit 41 determines the posting order of the submitted candidate articles and the submitted articles on the display surface visually recognized by the user. The ranking determination unit 41 searches for an article vector having a small inter-vector distance with each user vector of the user, and determines, for example, an article posting order in ascending order of the inter-vector distance. As a result, articles that are close to the user's interests and interests will be posted in a higher ranking. At this time, the ranking determination unit 41 may perform a search using a method similar to k-Nearest neighbor search.

In the ranking of articles determined by the ranking determination unit 41, the calculation unit 43 determines that the candidate article for submission is the first article (the article closest to the content that the user is interested in and the article at the top). The number of users (hereinafter referred to as "effective user number") is calculated. That is, the calculation unit 43 calculates information (for example, the publication rate) indicating what percentage of the user's ranking candidate articles can be ranked first. The calculation unit 43 calculates an index value that predicts the degree of interest of the user in the submission candidate article based on the article vector and the user vector.

The screen control unit 45 controls the screen displayed on the display unit 35. The screen control unit 45 is, for example, a screen (for display) including information on the submission candidate article input via the reception unit 33 and information on the number of users whose posting order of the submission candidate article is the highest. Information) is displayed on the display unit 35.

Each functional unit of the drafting device 30 is realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program (software). In addition, some or all of these components are hardware (circuits) such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), etc. It may be realized by the part; including circuitry), or it may be realized by the cooperation of software and hardware.

The storage unit 47 stores, for example, sample user information 49, article vector information 51, and user vector information 53. The sample user information 49 includes information of a predetermined number of users sampled from the users who have used the service provided by the service providing device 20. The sample user information 49 includes, for example, information on the user ID of the user sampled from the user included in the user information 22 of the service providing device 20.

The storage unit 47 is realized by, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), a flash memory, or a hybrid storage device in which a plurality of these is combined. Further, the storage unit 47 may be an external device such as NAS or an external storage server that can be accessed by the submission device 30.

[Operation of trafficking device]
Hereinafter, the operation of the submission device 30 will be described. FIG. 8 is a flowchart showing an example of the processing flow of the submission device 30. In the following, a predetermined number (for example, 1,000 people) sampled from the users included in the user information 22 of the service providing device 20 in the sample user information 49 of the storage unit 47 of the uploading device 30. It is assumed that the user information is stored in advance. Further, it is assumed that the article vector of the submitted article is stored in advance in the article vector information 51 of the storage unit 47 of the submission device 30.

First, the submission device 30 accepts the input of the submission candidate article input by the editor via the reception unit 33 (S101). The submission device 30 accepts, for example, input of text data of a submission candidate article.

Next, the submission device 30 generates an article vector corresponding to the input submission candidate article (S103). The trafficking device 30 generates an article vector by, for example, extracting words included in a trafficking candidate article by morphological analysis or the like, and then inputting a word vector distributed and expressed by a method such as word2vec into the first model. do.

Next, the submission device 30 extracts, for example, one user (hereinafter, referred to as “first sample user”) from the sample user information 49 stored in the storage unit 47 (S105), and the extracted user. The action log related to the above is acquired from the action log 26 of the service providing device 20 (S107).

Next, the trafficking device 30 reads an article vector corresponding to at least one article ID included in the acquired action log from the article vector information 51, and based on the read article vector, the user corresponding to the first sample user. Generate a vector. The trafficking device 30 generates a user vector by inputting, for example, a set of article vectors corresponding to at least one article ID included in the acquired action log into the second model (S109).

Next, the trafficking device 30 is based on the generated user vector corresponding to the first sample user, and when the trafficked article and the trafficking candidate article are targeted for distribution, the article to be included in the web page. The order of publication is determined (S111). For example, the trafficking device 30 searches for an article vector having a small inter-vector distance from the user vector corresponding to the first sample user, and determines the article posting order in ascending order of the inter-vector distance.

Next, the submission device 30 determines whether or not the submission candidate article is determined as the first article based on the determined publication order (S113). When the submission device 30 determines that the submission candidate article is determined as the first article, the submission device 30 increments the count value of the number of effective users by one (S115). On the other hand, when the submission device 30 determines that the submission candidate article is not determined as the first article, the submission device 30 does not perform the above increment processing.

Next, the submission device 30 determines whether or not the processing for all the sample users included in the sample user information 49 stored in the storage unit 47 is completed (S117). When the submission device 30 determines that the processing for all the sample users has not been completed, the submission device 30 returns to the processing of S105, extracts the unprocessed sample users, and performs the subsequent processing in the same manner.

On the other hand, when it is determined that the processing for all the sample users is completed, the submission device 30 generates a submission page including information on the submission candidate article and information on the number of effective users of the submission candidate article. It is displayed on the display unit 35 (S119).

FIG. 9 is a diagram showing an example of the submitted page P1. On this submission page P1, the titles of the five articles input as submission candidate articles in the submission device 30, the number of effective users, and the submission button "Submit" for accepting submission instructions are displayed. ing. By referring to this submission page P1, the editor can see how many effective users are when submitting a submission candidate article, that is, how many users the submission candidate article is delivered to. It becomes possible to grasp whether or not.

On this submission page P1, the editor presses the submission button of the submission candidate article that is finally submitted as a distribution article, performs editing work as appropriate, and then provides information including the article ID and article data. By transmitting to the providing device 20, the submission candidate article can be registered in the distribution article information 24 of the service providing device 20. After that, the submission candidate articles will be posted on the web page or the like provided by the service providing device 20.

According to the submission device of the embodiment described above, it is possible to appropriately assist the content registration work. In particular, when an editor performs editing work such as rewriting an article created in an external medium into a short summary and uploading it, a valid user of each submission candidate article before performing this editing work. You can figure out the number. Therefore, for example, the editor can prioritize the submission of articles delivered to many users, and can avoid unnecessary submission work for articles that are difficult to deliver to users.

In the above embodiment, an example in which the submission device 30 counts and displays the number of effective users has been described, but the submission candidate articles are determined as articles up to a predetermined rank (for example, the top 10). The number of users may be counted and displayed.

Further, the submission device 30 may calculate and display the degree of change in the occupancy rate, which indicates the ratio of the number of effective users among all the sample users, in addition to or instead of the number of effective users. For example, as shown in FIG. 10, the submission device 30 is used when the submission candidate article is not submitted due to the occupancy rate of the submission candidate article when the submitted article and the submission candidate article are targeted for distribution. The value obtained by subtracting the occupancy rate of the article determined as the first article among the submitted articles may be displayed on the trafficking page P2.

Further, in the above embodiment, an example in which the submission device 30 uses the sample user information 49 stored in advance in the device has been described, but instead of the sample user information 49, the service is provided every time the submission work is performed. User information may be acquired from the user information 22 of the providing device 20.

The information processing apparatus according to the embodiment of the present invention includes a first generation unit that generates first feature information corresponding to a first content that is a candidate to be distributed via a network, and at least one distributed via a network. The second generation unit that generates the second feature information corresponding to the user based on the user's access history to the second content, and the first feature information of the user and the second feature information based on the first feature information and the second feature information. (1) By providing a calculation unit that calculates an index value that predicts the degree of interest in the content, it is possible to appropriately assist the content registration work.

Although the embodiments for carrying out the present invention have been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. Can be added.

10 ... Terminal device 20 ... Service providing device 30 ... Drafting device 31 ... Communication unit 33 ... Reception unit 35 ... Display unit 37 ... Article vector generation unit 39 ... User vector generation unit 41 ... Ranking determination unit 43 ... Calculation unit 45 ... Screen Control unit 47 ... Storage unit

Claims (8)

The first generation unit that generates the first feature information corresponding to the first content that is a candidate to be distributed via the network, and the first generation unit.
A second generation unit that generates second feature information corresponding to the user based on the user's access history to at least one second content distributed via the network.
A determination unit that determines the placement order of the first content and the second content on the display surface visually recognized by the user based on the first feature information and the second feature information.
In determined the ad position, a calculation unit ranking of the first content is the number of users is within the predetermined rank is calculated as an index value predicted interest degree for the first content of the user,
Information processing device equipped with. A screen control unit for displaying a screen including information on the index value calculated by the calculation unit on the display device is further provided.
The information processing device according to claim 1. The calculation unit calculates the number of users having the highest ranking of the first content as the index value.
The information processing device according to claim 1 or 2. The calculation unit calculates, as the index value, a change in the ratio of the number of users who have the highest ranking of the first content among all the sample users.
The information processing device according to claim 3. Each of the first feature information and the second feature information is vector data.
The information processing device according to any one of claims 1 to 4. The first content is a candidate article to be distributed via the network.
The first feature information is information indicating the feature of the article, and is
The second content is an article distributed via the network.
The second feature information is information distributed via the network and derived based on the features of the article accessed by the user.
The information processing device according to any one of claims 1 to 5. The computer
Generates the first feature information corresponding to the first content that is a candidate to be distributed via the network,
Based on the user's access history to at least one second content distributed via the network, the second feature information corresponding to the user is generated.
Based on the first feature information and the second feature information, the placement order of the first content and the second content on the display surface visually recognized by the user is determined.
In the determined ranking, the number of users whose first content is within the predetermined ranking is calculated as an index value for predicting the degree of interest of the user in the first content.
Information processing method. On the computer
Generate the first feature information corresponding to the first content that is a candidate to be distributed via the network,
Based on the user's access history to at least one second content distributed via the network, the second feature information corresponding to the user is generated.
Based on the first feature information and the second feature information, the placement order of the first content and the second content on the display surface visually recognized by the user is determined.
In the determined ranking, the number of users whose first content is within the predetermined ranking is calculated as an index value for predicting the degree of interest of the user in the first content.
program.

Images