JP7830197B2 - Viewing information processing device, viewing information processing method, and viewing information processing program - Google Patents

Description

This invention relates to a viewing information processing device, a viewing information processing method, and a viewing information processing program, all of which have a function for selecting viewers from among viewers of a television device capable of outputting viewing log information, in order to investigate the viewing status of television broadcasts.

In recent years, television devices equipped with the ability to transmit viewing data, including viewing log information indicating which television channels were watched and at what time, to television manufacturers and other external networks such as the internet, have become widespread. In this specification, a device capable of receiving and viewing television broadcasts, as well as having the function of transmitting viewing log information as described above, is referred to as a television device.

Furthermore, as seen in Patent Document 1, for example, the applicant has proposed a technique for estimating, using a pre-machine learning-based model (mathematical model), the attribute composition of viewers of a television device (more specifically, one or more viewers who can watch television broadcasts received through the television device) and attribute-specific viewing information indicating the viewing status of television broadcasts for each attribute of the viewers, based on viewing log information obtained from each television device.

Here, the attribute composition of a television device's viewer refers to the composition of that viewer among multiple attribute categories, such as gender and age. Furthermore, the attribute-specific viewing information of a television device's viewer refers to whether or not viewers of each attribute watched a television broadcast on a particular channel (or broadcasting network) at a particular time and date.

Patent No. 6433615

Furthermore, when collecting viewing log information transmitted from television devices in a survey area where viewing patterns are to be investigated, the collected viewing log information pertains to viewers in households that own television devices connected to an external network. It does not include viewing log information for viewers who do not own such television devices or who do not use an external network. Therefore, if a survey of viewing patterns, such as viewership ratings, is conducted using the entirety of the collected viewing log information, the estimated viewing patterns may deviate significantly from the actual viewing patterns in the survey area.

For example, viewers who watch television broadcasts on television devices that have communication connections to external networks to transmit viewing log information are often from a relatively young generation. Furthermore, this generation tends to spend a significant portion of their day outside the home, such as at work or school, or working or studying at home, resulting in a relatively short amount of time spent watching television broadcasts.

On the other hand, while viewers with attributes such as the elderly often spend a long time watching television broadcasts each day, they often do not own a television device with a function to transmit viewing log information, or they do not utilize a communication connection between such a television device and an external network.

In this situation, when collecting viewing log information transmitted from television devices, the overall collection of viewing log information tends to be dominated by younger generations (those with relatively short viewing times), while viewing log information for older generations and others with relatively longer viewing times will be less abundant compared to that of younger generations. Therefore, there is a risk that the results of surveys on viewing patterns estimated from the collected viewing log information may deviate from the actual viewing patterns in the surveyed area.

This invention was made in view of the above background, and aims to provide a viewing information processing device and a viewing information processing method that can select viewers to be surveyed in order to appropriately identify the television viewing situation in a surveyed area from viewing log information acquired from multiple television devices in the surveyed area. Furthermore, it aims to provide a viewing information processing program that allows a computer to perform the selection of such viewers.

To achieve the above objective, the present invention provides a plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channels or broadcasting station networks were viewed through the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, a viewing time acquisition unit acquires the viewing time of television broadcasts within a predetermined unit time interval for each attribute included in the viewer attribute configuration of each television device,
A frequency distribution identification unit identifies the frequency distribution of the number of viewers relative to viewing time for each attribute in the survey area, based on the viewing time for each attribute estimated by the viewing time estimation unit.
The system includes a target viewer selection unit that, based on a comparison between the frequency distribution of the number of viewers for each attribute identified by the frequency distribution identification unit and the frequency distribution of predetermined criteria for each attribute in the survey area, selects viewers to be surveyed for each attribute in the survey area to determine the viewing status in the survey area.
The viewer selection unit for the survey is configured to select viewers for each attribute such that the frequency distribution of the number of viewers for the survey relative to viewing time matches the frequency distribution of the criteria (first invention).

To elaborate, in this invention, the attribute configuration of viewers for each television device refers to the overall composition of viewers for each television device, based on their respective attributes.
Furthermore, the frequency distribution of the number of viewers surveyed in relation to viewing time (hereinafter sometimes simply referred to as Frequency Distribution A in this column) matching the reference frequency distribution means that the pattern of change in the frequency (number of viewers) according to viewing time in Frequency Distribution A matches the pattern of change in the reference frequency distribution. In this case, "matching" includes not only cases where they match exactly, but also cases where they match almost (approximately). In addition, as an indicator for judging the degree of matching of the above "change patterns," for example, the ratio of the frequencies corresponding to each of several types of viewing time can be used.

According to the present invention, by comprising a viewing time acquisition unit and a frequency distribution identification unit, it is possible to identify the frequency distribution of the number of viewers relative to viewing time for each attribute in the surveyed area, based on the viewing time of viewers for each attribute of each television device (viewing time of television broadcasts within a predetermined unit time interval).

Then, the audience selection unit selects audience members for each attribute such that the frequency distribution of the number of audience members relative to viewing time matches the baseline frequency distribution. In this case, by appropriately setting the baseline frequency distribution for each attribute, it is possible to select audience members such that the frequency distribution of the number of audience members for each attribute reflects the actual situation in the surveyed area.

Therefore, by conducting a viewing survey on the viewers selected as described above, it becomes possible to obtain survey results that match the actual viewing situation in the surveyed area. Thus, according to the present invention, it is possible to select target viewers in a way that appropriately identifies the television viewing situation in the surveyed area from viewing log information acquired from multiple television devices in the surveyed area.

In the first invention described above, the frequency distribution of the standard may be set based on viewing data obtained from each of a plurality of sample households pre-selected in the survey area, and which includes viewing log information for each viewer in each sample household (second invention).

According to this, by appropriately selecting sample households, it is possible to set a suitable frequency distribution for selecting viewers for each attribute, so that the frequency distribution of the number of viewers surveyed matches the actual situation in the surveyed area.

The first or second invention may further include a viewing status estimation unit that estimates the viewing status of a predetermined television broadcast to be investigated, and the viewing status estimation unit has the function of estimating whether or not each target viewer selected by the target viewer selection unit has watched the television broadcast to be investigated, based on viewing log information acquired by the viewing log information acquisition unit with respect to the television device corresponding to the viewer, and is configured to estimate the viewing status of the television broadcast to be investigated for each attribute of the target viewer based on the results of the estimation for all target viewers in the target area (third invention).

According to this method, based on viewing log information obtained from television devices corresponding to the selected target audiences as described above, it is possible to estimate whether or not each target audience watched the target television broadcast. Based on this estimation, it becomes possible to estimate the viewing status of the target television broadcast in the target area with high reliability for each viewer attribute.

Furthermore, the present invention relates to a plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channel or broadcasting station's television broadcast was viewed through the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, the first step is to acquire the viewing time of television broadcasts within a predetermined unit time interval for each attribute included in the viewer attribute configuration of each television device,
A second step involves identifying the frequency distribution of the number of viewers relative to viewing time for each attribute in the survey area, based on the viewing time for each attribute obtained in the first step.
The system includes a third step in which, based on a comparison between the frequency distribution of the number of viewers for each attribute identified in the second step and the frequency distribution of predetermined criteria for each attribute in the survey area, viewers to be surveyed for each attribute in the survey area are selected for the survey.
The third step is characterized in that, for each attribute, the viewers to be surveyed are selected such that the frequency distribution of the number of viewers to be surveyed relative to viewing time matches the frequency distribution of the criteria (fourth invention).
According to this fourth invention, the same effects as the first invention can be achieved.

Furthermore, the present invention provides a plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channels or broadcasting station networks were viewed via the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, when the viewing time of television broadcasts within a predetermined unit time interval is given for each attribute included in the viewer attribute configuration of each television device, the first process identifies the frequency distribution of the number of viewers with respect to viewing time for each attribute in the survey area based on the viewing time.
The system is configured to perform a second process on the computer, which involves comparing the frequency distribution of the number of viewers for each attribute identified by the first process with the frequency distribution of predetermined criteria for each attribute in the survey area, and then selecting viewers to be surveyed for each attribute in the survey area.
The second process is characterized in that it causes the computer to select viewers for each attribute such that the frequency distribution of the number of viewers to be surveyed relative to viewing time matches the frequency distribution of the criteria (Fifth Invention).
According to this fifth invention, the same effects as the first invention can be achieved by having a computer execute the visual information processing program.

A diagram showing the overall system in an embodiment of the present invention. A block diagram illustrating the processing of the attribute configuration estimation unit of the viewing information processing device shown in Figure 1. A block diagram illustrating the processing of the viewing time estimation unit of the viewing information processing device shown in Figure 1. A graph illustrating the processing of the frequency distribution identification unit and the target audience selection unit shown in Figure 1.

One embodiment of the present invention will be described below with reference to Figures 1 to 4. Referring to Figure 1, the system described in this embodiment includes a viewing information processing device 1 that performs processing related to the investigation of television broadcast viewing status. The viewing information processing device 1 is composed of, for example, one or more computers. These computers include a processor such as a CPU (not shown), memory (storage device), interface circuits, communication devices, etc. The viewing information processing device 1 can acquire viewing data related to television broadcast viewing in each device-specific household 20 from multiple device-specific households 20 belonging to a target area for the investigation of viewing status (for example, a prefecture-level area, or an area combining multiple prefectures such as the Kanto region or the Kinki region). This viewing data includes viewing log information indicating which television broadcasts (or which broadcasting station network) were viewed and at what time.

In this embodiment, each "device-specific household 20" refers to a household comprising a television device 23 that includes a television broadcast receiver 21 and a viewing data output device 22 capable of outputting viewing data related to television broadcast viewing, and whose members include one or more viewers of the television device 23 (specifically, one or more viewers capable of viewing television broadcasts received by the receiver 21 of the television device 23). In other words, a device-specific household 20 refers to a household configured as a pair of a television device 23 and its viewers.

In this case, one device-specific household 20 is associated with each individual television device 23. Therefore, even if a dwelling unit is equipped with multiple television devices 23, one device-specific household 20 is associated with each of those multiple television devices 23. In this case, each device-specific household 20 corresponding to each of the multiple television devices 23 may include the same members as viewers. In other words, one or more members of one device-specific household 20 may also be members of another device-specific household 20.

Each device-specific household 20's television device 23 may consist of a television, or a television and a recording device connected thereto. The viewing data output device 22 of the television device 23 is composed of, for example, a processor such as a microcontroller (not shown), memory, interface circuits, communication devices, etc. The viewing data output device 22 is capable of detecting the channels of television broadcasts viewed through the television device 23, and is also capable of detecting the date and time the television broadcast of that channel was viewed (specifically, the start and end dates and times of viewing the television broadcast of that channel). From this detection information, it is possible to generate viewing log information for the device-specific household 20 (hereinafter referred to as device-specific viewing log information). In other words, this device-specific viewing log information is the overall viewing log information for all members included in the device-specific household 20.

The viewing data output device 22 can periodically (or upon request from the server) transmit viewing data, including the generated device-specific viewing log information, to a server (not shown) of the manufacturer of the television or recording device constituting the television device 23 via an external network NW, such as the Internet or a telephone network. In this case, the viewing data transmitted from the viewing data output device 22 includes, in addition to the device-specific viewing log information, location information (for example, information indicating the first three digits of the postal code) pre-registered on the television or recording device constituting the television device 23, and identification information of the television device 23. The identification information of the television device 23 can also be used as identification information for the device-specific household 20 that owns the television device 23.

The viewing information processing device 1 comprises a viewing data acquisition unit 11, an attribute configuration estimation unit 12, a viewing time acquisition unit 13, a frequency distribution identification unit 14, a target viewer selection unit 15, and a viewing status estimation unit 16, all of which are realized through the implemented hardware configuration and program (software configuration). The overall processing of the viewing information processing device 1 will be described below, along with details of each of these functional units.

The viewing data acquisition unit 11 is capable of communicating with the manufacturer's server (not shown) of each television device 23 in each device-specific household 20. By communicating with this server, it can acquire viewing data for each television device 23 of that manufacturer. This viewing data acquisition unit 11 functions as a viewing log information acquisition unit in this invention.

Furthermore, if the viewing information processing device 1 can communicate with the viewing data output device 22 of each device-specific household 20, the viewing data acquisition unit 11 may directly acquire viewing data from the television device 23 of each device-specific household 20 from the viewing data output device 22 of the television device 23. Alternatively, the viewing information processing device 1 may acquire viewing data from each manufacturer via an appropriate storage device.

In this embodiment, the viewing information processing device 1 sequentially executes the processing of the attribute configuration estimation unit 12 and the viewing time information acquisition unit 13 for each household 20 of each device type in the survey area, using the viewing data acquired by the viewing data acquisition unit 11.

The attribute configuration estimation unit 12 is a functional unit that estimates the attribute configuration (in other words, the attribute configuration of viewers of the television device 23 in each device-specific household 20) from device-specific viewing log information for a predetermined period (e.g., one month, several months, one year, etc.) output from the television device 23 of each device-specific household 20.

Here, the attributes of the members of each device-specific household 20 (viewers of the television device 23) are classified into multiple attribute types, for example, according to the age and gender of the members. For example, they may be classified into multiple attribute types such as children under y1, males aged y1 and over but under y2, females aged y1 and over but under y2, males aged y2 and over, and females aged y2 and over. Hereafter, the number of attribute types will be N, and each of these N attribute types will be appropriately denoted as At(i) (i = 1, 2, ..., N). Note that the attributes of the members of each device-specific household 20 can be classified not only according to age and gender, but also according to various parameters such as occupation and educational background.

As shown in Figure 2, the attribute composition estimation unit 12 estimates the attribute composition of the members of each device-specific household 20 using a first model that has been pre-processed with machine learning. Here, the first model is a model that has been pre-processed with machine learning so that it can estimate the attribute composition of the members of any device-specific household 20 belonging to the survey area from device-specific viewing logs for a predetermined period of time of said device-specific household 20. In the machine learning processing of the first model, viewing log information and attribute composition information for each sample household, obtained from people meters installed in multiple sample households with known attribute compositions, are used as training data. Furthermore, a known algorithm can be used as the algorithm for this machine learning processing.

In this embodiment, the first model is configured to identify the attribute-specific existence probability, which is the probability that a component of each attribute At(i) exists in each device-specific household 20, for each of the N types of attribute At(i) (i = 1, 2, ..., N), based on device-specific viewing log information for a predetermined period of time for each device-specific household 20. As such a first model, for example, a model similar to the mathematical model for determining household composition in Patent Document 1 can be adopted. However, the first model may be any other form of model, as long as it can identify the attribute-specific existence probability (or a similar index value) from device-specific viewing log information for a predetermined period of time for each device-specific household 20.

The attribute configuration estimation unit 12 then estimates the attributes of the members of the device-specific household 20 based on attributes whose probability of existence, as identified by the first model, is higher than a predetermined threshold (e.g., 0.5) (or attributes whose probability of existence is higher than or equal to the predetermined threshold). As an example, Figure 2 illustrates the processing of the attribute configuration estimation unit 12 for one device-specific household 20 (in Figure 2, the device-specific household 20 with identification information ID x1). In the block diagrams shown in Figure 2 and Figure 3 (described later), the processing execution unit is indicated by a thick line frame, and the data output unit or data is indicated by a thin line frame.

In the example shown in Figure 2, the first model identifies the attribute-specific existence probabilities of the members of a device-specific household 20, whose identification information ID is x1, from the device-specific viewing log information for a predetermined period, as shown in the figure. In this example, among the N types of attributes At(i) (i = 1, 2, ..., N), the attribute-specific existence probabilities of attributes At(1), At(3), and At(N) are higher than a predetermined threshold (here, for example, 0.5), while the attribute-specific existence probabilities of the other attributes At(2), At(4) to At(N-1) are lower than this threshold.

In this case, the attribute configuration estimation unit 12 estimates that attributes At(1), At(3), and At(N), whose probability of existence is higher than the threshold, are attributes of the members of the device-specific household 20 (viewers of the television device 23), and attributes At(2), At(4) to At(N-1), whose probability of existence is lower than the threshold, are not attributes of the members of the device-specific household 20. This allows the attribute configuration of the members of the device-specific household 20 to be estimated.

In this way, the attribute composition estimation unit 12 estimates the attribute composition of the members of each device-specific household 20 by estimating whether or not members (viewers) of each attribute exist in that device-specific household 20 based on the relative levels of the attribute-specific existence probabilities identified by the first model relative to predetermined thresholds.

The viewing information processing device 1 executes the processing of the attribute configuration estimation unit 12, and then executes the processing of the viewing time acquisition unit 13. In this embodiment, the viewing time acquisition unit 13 is a functional unit that, for each household 20 per device, estimates and acquires the viewing time of television broadcasts by viewers of each attribute within a predetermined unit time interval, for example, the daily viewing time, for each attribute included in the attribute configuration estimated by the attribute configuration estimation unit 12. Note that the above unit time interval is not limited to one day, but may be one week, one month, or other time intervals.

In this case, the viewing time acquisition unit 13 includes, for each device-specific household 20, an attribute-specific viewing information generation unit 13a that generates (estimates) attribute-specific viewing information using a second model, indicating whether or not viewers of each attribute included in the attribute configuration of the device-specific household 20 watched television broadcasts corresponding to any combination of channel (or broadcasting station network) and date/time data indicating month, day of the week, time slot, etc., based on the viewing log information of each device-specific household 20; and a viewing time aggregation unit 13b that aggregates the viewing time (total viewing time over a predetermined unit time interval) for each attribute included in the attribute configuration of the device-specific household 20 from the attribute-specific viewing information.

Furthermore, a television broadcast corresponding to a pair of channel (or broadcasting network) and date/time data indicating the month, day of the week, and time slot refers to a television broadcast aired on that channel (or broadcasting network) at the date and time indicated by that date/time data. The time slot within the date/time data is, for example, a time slot divided into predetermined time units (10 minutes, 30 minutes, 1 hour, etc.).

Here, the second model used in the attribute-specific viewing information generation unit 13a is a model created separately for each of the N types of attributes At(i) (i = 1, 2, ..., N). The second model corresponding to each attribute At(i) is a model that has undergone pre-machine learning processing so that, when a set of channel (or broadcasting station network) and date/time data is specified, it can identify the probability (hereinafter referred to as attribute-specific viewing probability) that a viewer of attribute At(i) watched a television broadcast corresponding to the specified channel (or broadcasting station network) and date/time data set (hereinafter referred to as channel- and date/time specified broadcast) from the device-specific viewing logs of the device-specific household 20 to which the constituent members (viewers) of attribute At(i) belong for a predetermined period.

Figure 2 shows that, based on viewing log information for device-specific households 20 with a component (viewer) of attribute At(n), the attribute-specific viewing probability—the probability that a component with attribute At(n) viewed a channel and time-specified broadcast—can be identified by a second model corresponding to attribute At(n). The same applies to the second model corresponding to other attributes.

In the machine learning processing for the second model described above, the viewing log information for each member of each sample household, obtained from people meters installed in multiple sample households with known attribute configurations, and the attribute configuration of each sample household are used as training data. A known algorithm can be used as the algorithm for this machine learning processing. Furthermore, such a second model can be similar to, for example, the mathematical model for individual viewing determination described in Patent Document 1. However, the second model may be of any other form, as long as it can identify attribute-specific viewing probabilities (or similar indicator values) for channel (or broadcasting station network) and date/time specified broadcasts, based on the device-specific viewing log information for a predetermined period of time for each member of the device-specific household 20.

Furthermore, the attribute-specific viewing information generation unit 13a is configured to determine (estimate) whether or not a member of each attribute has viewed a channel- and time-specified broadcast, based on the attribute-specific viewing information estimated by the second model and the device-specific viewing log information of the device-specific household 20, for each attribute that constitutes the attribute configuration of the device-specific household 20.

Specifically, the attribute-specific viewing information generation unit 13a determines that a member of a household 20 by device has viewed a channel-specific broadcast if the attribute-specific viewing probability value (probability value) identified by the second model for channel-specific broadcasts is above a predetermined threshold (for example, 0.5 or higher), and if it is confirmed from the viewing log information of the household 20 by device that actual viewing occurred on the channel (or broadcasting station network) of the channel-specific broadcast at the broadcast date and time indicated by the date and time data of the channel-specific broadcast. If these conditions are not met, the unit determines that there was no channel-specific broadcast by the member of that attribute.

The viewing time acquisition unit 13 sequentially specifies each of the channel (or broadcasting station network) and date/time data sets for each attribute included in the attribute configuration of each device-specific household 20 to the attribute-specific viewing information generation unit 13a configured as described above, within a predetermined period (for example, within a one-month period). It then estimates whether the channel/date/time specified broadcast corresponding to each of these sets was viewed from the device-specific viewing logs for the predetermined period of the device-specific household 20.

Next, the viewing time acquisition unit 13 executes the processing of the viewing time aggregation unit 13b. In this viewing time aggregation unit 13b, for each attribute of the members of each household 20 for each device, the sum of the time intervals of all channels and scheduled broadcasts that are estimated to have been viewed is calculated, and the daily viewing time is estimated by dividing this sum by the number of days within the predetermined period. In this way, the daily viewing time is obtained for each attribute of the members (viewers) of each household 20 for each device.
In this embodiment, the processing of the viewing time estimation unit 13 is performed as described above.

The viewing information processing device 1 performs the processing of the attribute configuration estimation unit 12 and the viewing time acquisition unit 13 as described above for each device-specific household 20 in the survey area, and then performs the processing of the frequency distribution identification unit 14. This frequency distribution identification unit 14 is a functional unit that identifies the frequency distribution of viewing time (daily viewing time) in the survey area (more specifically, the frequency distribution of the number of viewers relative to viewing time) for each of the N types of attributes At(i) (i = 1, 2, ..., N). In this case, the viewing time is divided into m types of time ranges T(k) (k = 1, 2, ..., m). For example, it is divided into m types of time ranges such as T(1) = 0, 0 < T(2) ≤ Tx_2, Tx_2 < T(3) ≤ Tx_3, ..., Tx_m-2 < T(m-1) ≤ Tx_m-1, Tx_m-1 < T(m). The boundary values Tx_2, Tx_3, ..., Tx_m-1 for adjacent time ranges are predetermined constant values.

The frequency distribution identification unit 14 then determines, for each of the N types of attribute At(i) (i = 1, 2, ..., N), the total number of viewers (total number in the survey area) whose viewing time belongs to each time range T(k) as the frequency corresponding to that time range T(k). For example, if, in the entirety of device-specific households 20 estimated to have viewers (components) of a certain attribute At(n), the total number of viewers (components of attribute At(n)) whose viewing time belongs to T(1), T(2), ..., T(m) is X1, X2, ..., Xm, respectively, then X1, X2, ..., Xm are determined as the frequencies corresponding to the viewing time in each time range T(1), T(2), ..., T(m) for attribute At(n), respectively.

The frequency distribution identification unit 14 identifies the frequency distribution related to viewing time for each of the N types of attributes At(i) (i = 1, 2, ..., N) in this manner. The solid bar graph in Figure 4 illustrates the frequency distribution of viewing time identified as described above for one of the attributes At(i). In this example, the number of types m in the time range T(k) is, for example, 10. The frequency distribution of viewing time is similarly identified for the other attributes.

To elaborate, the number of types m in the time range T(k), or the boundary values Tx_2, Tx_3, ..., Tx_m-1 of adjacent time ranges, do not need to be the same for all N types of attributes At(i) (i = 1, 2, ..., N). They may differ depending on the type of attribute At(i). For example, the number of types m in the time range T(k) corresponding to attributes where viewing time is easily distributed across a wide range may be greater than the number of types m in the time range T(k) corresponding to attributes where viewing time is relatively less distributed.

The viewing information processing device 1 then executes the processing of the target viewer selection unit 15. This target viewer selection unit 15 is a functional unit that selects target viewers—those who will be used to investigate viewing conditions such as television broadcast viewership ratings—for each of the N types of attribute At(i) (i = 1, 2, ..., N).

In this case, the target audience selection unit 15 selects target audiences for each of the N types of attribute At(i) (i = 1, 2, ..., N) based on a comparison between the frequency distribution identified by the frequency distribution identification unit 14 and the predetermined standard frequency distribution corresponding to each attribute At(i).

Specifically, in this embodiment, the frequency distribution of viewing time for each attribute At(i) within the sample households is predetermined from the viewing log information of each member of each sample household obtained via a people meter or the like from multiple sample households (households where the attributes of each member are known) selected in advance within the survey area. This frequency distribution is used as the frequency distribution for the above-mentioned criteria. Here, since the sample households are selected as households that represent the television broadcasting viewing situation in the survey area, the frequency distribution of viewing time for each attribute At(i) within the sample households can be considered to correspond to the actual frequency distribution of viewing time for each attribute At(i) in the survey area.

The target audience selection unit 15 then selects target audiences for each attribute At(i) such that the frequency distribution of the selected target audiences' viewing time matches (or nearly matches) the reference frequency distribution. Such selection can be performed for each attribute At(i) as follows, for example:

Hereafter, any one attribute from among the N types of attributes At(i) (i = 1, 2, ..., N) will be referred to as the focus attribute At(x), and the process for selecting target viewers with respect to this focus attribute At(x) will be explained below. Furthermore, in the following explanation, the frequency distribution of viewing time identified for each attribute At(i) by the frequency distribution identification unit 14 will be referred to as the pre-selection frequency distribution, the frequency in each time range T(k) (k = 1, 2, ..., m) within the pre-selection frequency distribution will be referred to as the pre-selection frequency Xa_k, the frequency distribution of viewing time of the target viewers selected for each attribute At(i) will be referred to as the post-selection frequency distribution, and the frequency in each time range T(k) (k = 1, 2, ..., m) within the post-selection frequency distribution will be referred to as the post-selection frequency Xb_k. Furthermore, in the reference frequency distribution, the frequency (normalized frequency) corresponding to each time range T(k) (k = 1, 2, ..., m) is called the reference frequency Xs_k. As an example of the pre-selection frequency distribution for the above-mentioned attribute At(x), we use the frequency distribution shown by the solid bar graph in Figure 4. As an example of the reference frequency distribution corresponding to the above-mentioned attribute At(x), we use the frequency distribution shown by the dashed line with the white circles as inflection points in Figure 4.

(Step 1)
The target audience selection unit 15 calculates the value Xa_k/Xs_k (hereinafter referred to as the frequency ratio to the reference) by dividing the pre-selection frequency Xa_k corresponding to each of the m (in this case, 10) time ranges T(1) to T(10) in the pre-selection frequency distribution of audiences with the attribute of interest At(x) by the reference frequency Xs_k corresponding to each of the m time ranges T(1) to T(10) in the reference frequency distribution. Then, from the m (10) time ranges T(1) to T(10), it identifies the time range corresponding to the smallest frequency ratio to the reference Xa_k/Xs_k (excluding zero) (hereinafter referred to as the minimum frequency ratio time range T(min)). For example, in the example of the pre-selection frequency distribution (shown as a solid bar graph) and the reference distribution shown in Figure 4, among the pre-selection frequencies Xa_1 to Xa_10 corresponding to each of the 10 time ranges T(1) to T(10), the ratio of Xa_9/Xs_9 (>0) to the reference frequency corresponding to time range T(9) is the smallest, so T(9) is identified as the time range T(min) with the minimum frequency ratio.

(Step 2)
The target audience selection unit 15 calculates the reference ratio as the ratio of the frequency in each of the other time ranges to the frequency in the minimum frequency ratio time range T(min) in the reference frequency distribution corresponding to the attribute of interest At(x). For example, in the reference frequency distribution illustrated in Figure 4, the reference ratios corresponding to each time range T(1), T(2), ..., T(8), T(10) other than the minimum frequency ratio time range T(min) (=T(9)) are calculated by the division operations Xs_1/Xs_9, Xs_2/Xs_9, ..., Xs_8/Xs_9, Xs_10/Xs_9, respectively.

(Step 3)
The audience selection unit 15 sets the pre-selection frequency Xa_9 in the minimum frequency time range T(min) (=T(9)) as the provisional target value for the post-selection frequency Xb_9 in the minimum frequency ratio time range T(min), and sets the provisional target values for the post-selection frequencies Xb_1, Xb_2, ..., Xb_8, Xb_10 in each of the other time ranges T(1), T(2), ..., T(8), T(10) as the provisional target values for the post-selection frequencies Xb_1, Xb_2, ..., Xb_8, Xb_10.

Therefore, when any one of the time ranges T(1), T(2), ..., T(8), T(10) other than the minimum frequency ratio time range T(min) (=T(9)) is denoted as T(y) (where y is one of 1, 2, ..., 8, or 10), the integer part of the value obtained by multiplying the provisional target value of the post-selection frequency Xb_y corresponding to the time range T(y) by the reference ratio (=Xs_y/Xs_9) corresponding to the time range T(y) is set as the provisional target value of the post-selection frequency Xb_9 in the minimum frequency ratio time range T(min) (=T(9)) (=pre-selection frequency Xa_9).

(Step 4)
The audience selection unit 15 determines, for each time range T(k) (k = 1, 2, ..., 10), whether the provisional target value of the post-selection frequency Xb_k set in step 3 is less than or equal to the pre-selection frequency Xa_k corresponding to the time range T(k) (whether the provisional target value of the post-selection frequency Xb_k ≤ pre-selection frequency Xa_k). Note that, with respect to the minimum frequency ratio time range T(min) (= T(9)), the result of this determination process will inevitably be positive, so this determination process for the minimum frequency ratio time range T(min) may be omitted.

(Step 5)
The audience selection unit 15 determines the target value of the post-selection frequency Xb_k in each time range T(k) as the provisional target value of the post-selection frequency Xb_k if the judgment result in step 4 is positive for all time ranges T(1) to T(10). For example, in Figure 10, the judgment result in step 4 is positive for all time ranges T(1) to T(10). In this case, the provisional target value of the post-selection frequency Xb_k in each time range T(k) (k = 1, 2, ..., 10) is determined as the target value of the post-selection frequency Xb_k.

(Step 6)
The audience selection unit 15, if the judgment result in step 4 is negative for any one or more time ranges from time range T(1) to T(m), sets a new provisional target value for the post-selection frequency Xb_9 in the minimum frequency ratio time range T(min) (=T(9)) that is smaller than the pre-selection frequency Xa_9 in the same minimum frequency ratio time range T(min).

In this case, the new provisional target value for the post-selection frequency Xb_9 in the minimum frequency ratio time range T(min) (=T(9)) is set such that, when provisional target values for the post-selection frequencies in other time ranges are set using this new provisional target value through the same process as in step 3, the condition that the judgment result in step 4 becomes positive for all time ranges T(1) to T(10) is satisfied. Furthermore, the new provisional target value for the post-selection frequency Xb_9 in the minimum frequency ratio time range T(min) is set to be as close as possible to the pre-selection frequency Xa_9 in that minimum frequency ratio time range T(min) within the range that satisfies the above condition.

Setting a new provisional target value for the post-selection frequency Xa_9 within the minimum frequency ratio time range T(min) can be achieved, for example, by repeatedly decreasing the provisional target value from the pre-selection frequency Xa_9 by a predetermined amount until the above conditions are met.

Then, the audience selection unit 15, after setting new provisional target values for the post-selection frequency Xb_k in each time range T(k) to satisfy the above conditions, determines, in the same manner as in step 5, the provisional target values for the post-selection frequency Xb_k in each time range T(k) as the target values for the post-selection frequency Xb_k in each time range T(k).

(Step 7)
In step 5 or step 6, after determining the target value of the post-selection frequency Xb_k in each time range T(k), the target audience selection unit 15 removes a number of viewers (viewers with the focus attribute At(x)) from the total number of viewers who have viewing time belonging to each time range T(k) (k = 1, 2, ..., m), corresponding to the difference between the pre-selection frequency Xa_k and the target value of the post-selection frequency Xb_k for that time range T(k), and selects the remaining viewers as target audiences for the focus attribute At(x). In this case, the viewers to be removed in each time range T(k) are selected, for example, randomly. As a result, the target audiences (viewers with the focus attribute At(y)) for each time range T(k) (k = 1, 2, ..., m) are selected so that their total number matches the target value of the post-selection frequency Xb_k.

For example, in Figure 4, for each time range T(k) (k = 1, 2, ..., 10) where the pre-selection frequency is greater than the target post-selection frequency, the number of viewers corresponding to the frequency in the gray area is removed. As a result, the target viewers are selected so that the post-selection frequency distribution for viewing time for the focus attribute At(x) matches the reference frequency distribution. The same process is then used to select target viewers for other attributes. The selected target viewers are then stored in association with the television devices 23 of the device-specific households 20 to which they belong.
In this embodiment, the processing of the target audience selection unit 15 is performed as described above.

The viewing information processing device 1, after selecting target viewers for each attribute At(i) as described above, executes the processing of the viewing status estimation unit 16 using these target viewers. This viewing status estimation unit 16 is a processing unit that estimates viewing status data (data equivalent to viewership ratings) for each of the N types of attributes At(i) regarding the target television broadcast (hereinafter referred to as the target broadcast) in the target area.

In this case, the viewing status estimation unit 16 estimates whether or not a target viewer watched the target broadcast by performing the same processing as the attribute-based viewing information generation unit 13a of the viewing time estimation unit 13 (processing that uses a second model corresponding to the attributes of the target viewer) based on viewing log information for a predetermined period for each device-specific household 20 to which each target viewer belongs, and the channel (or broadcasting station network) and date/time data of the target broadcast.

The viewing status estimation unit 16 then estimates whether or not all target viewers in the survey area have watched the target broadcast, and based on the estimation results, calculates the viewing status data for each attribute At(i) of the target broadcast, for example, using the following formula (1).

Viewing status data related to attribute At(i) = m(i) / M(i) ... (1)

Here, M(i) is the total number of surveyed viewers with attribute At(i), and m(i) is the total number of surveyed viewers with attribute At(i) who are estimated to have watched the surveyed broadcast.

This allows us to obtain viewing status data equivalent to viewership ratings for each attribute A(i) in the surveyed area. In this case, since the surveyed viewers are selected so that the post-selection frequency distribution regarding viewing time matches the standard frequency distribution based on viewing data for sample households in the surveyed area, highly reliable viewing data can be obtained.

In the embodiment described above, for each time range T(k), the provisional target value of the post-selection frequency Xb_k, set as described above, was determined as the target value of the post-selection frequency Xb_k for each time range T(k), provided that the provisional target value of the post-selection frequency Xb_k ≤ the pre-selection frequency Xa_k for all time ranges T(1) to T(10).

However, even if the provisional target values of the post-selection frequencies corresponding to some of the time ranges T(1) to T(10) are greater than the pre-selection frequencies corresponding to those time ranges, if the difference between the provisional target values and the pre-selection frequencies for those time ranges is sufficiently small (below a predetermined threshold), the target values of the post-selection frequencies for those time ranges may be set to match the pre-selection frequencies, while the provisional target values of the post-selection frequencies may be set as the target values for the time ranges other than those time ranges (time ranges where the provisional target values of the post-selection frequencies are less than or equal to the pre-selection frequencies).

Furthermore, while the above embodiment included a viewing status estimation unit 16, the viewing information processing device in the present invention may not include the viewing status estimation unit 16. In this case, the viewing status estimation unit 16 may be provided in a separate device from the viewing information processing device.

Furthermore, in the above embodiment, the attribute configuration and attribute-specific viewing information of each device-specific household 20 were estimated from the device-specific viewing log information, and the viewing time for each attribute of viewers in each device-specific household 20 (viewing time within a predetermined unit time interval) was estimated from the estimated attribute-specific viewing information. However, in the present invention, the method for obtaining the viewing time for each attribute of viewers in each device-specific household 20 is not limited to the above method. For example, each device-specific household 20 may be equipped with an appropriate device capable of measuring or estimating the viewing time for each attribute of viewers in each device-specific household 20, and the viewing time for each attribute of viewers in each device-specific household 20 may be obtained from that device. Alternatively, for example, each device-specific household 20 may be equipped with an appropriate device capable of generating and outputting viewing log information for each attribute of viewers in each device-specific household 20, and the viewing time for each attribute of viewers in each device-specific household 20 may be estimated from the viewing log information output from that device.

1...Viewing information processing device, 13...Viewing time acquisition unit, 14...Frequency distribution identification unit, 15...Survey target viewer selection unit, 16...Viewing status estimation unit.

Claims (5)

A plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channels or broadcasting station networks were viewed through the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, a viewing time acquisition unit acquires the viewing time of television broadcasts within a predetermined unit time interval for each attribute included in the viewer attribute configuration of each television device,
A frequency distribution identification unit identifies the frequency distribution of the number of viewers relative to viewing time for each attribute in the survey area, based on the viewing time for each attribute obtained by the viewing time acquisition unit.
The system includes a target viewer selection unit that, based on a comparison between the frequency distribution of the number of viewers for each attribute identified by the frequency distribution identification unit and the frequency distribution of predetermined criteria for each attribute in the survey area, selects viewers to be surveyed for each attribute in the survey area to determine the viewing status in the survey area.
The viewing information processing device is characterized in that the viewer selection unit for the survey is configured to select viewers for each attribute such that the frequency distribution of the number of viewers for the survey relative to viewing time matches the frequency distribution of the criteria. In the viewing information processing device according to claim 1,
The frequency distribution of the aforementioned criteria is a frequency distribution set based on viewing data obtained from each of a plurality of sample households pre-selected in the survey area, and which includes viewing log information for each viewer in each sample household. In the viewing information processing device according to claim 1 or 2,
The viewing information processing device further comprises a viewing status estimation unit for estimating the viewing status of a predetermined target television broadcast, the viewing status estimation unit having the function of estimating whether each target viewer selected by the target viewer selection unit has watched the target television broadcast, based on viewing log information acquired by the viewing log information acquisition unit with respect to the television device corresponding to the viewer, and is configured to estimate the viewing status of the target television broadcast for each attribute of the target viewer based on the results of the estimation for all target viewers in the target area. A plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channels or broadcasting station networks were viewed through the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, the first step is to obtain the viewing time of television broadcasts within a predetermined unit time interval for each attribute included in the viewer attribute configuration of each television device,
A second step involves identifying the frequency distribution of the number of viewers relative to viewing time for each attribute in the survey area, based on the viewing time for each attribute obtained in the first step.
The system includes a third step in which, based on a comparison between the frequency distribution of the number of viewers for each attribute identified in the second step and the frequency distribution of predetermined criteria for each attribute in the survey area, viewers to be surveyed for each attribute in the survey area are selected for the survey.
The third step is a method for processing viewing information, characterized in that, for each attribute, the frequency distribution of the number of viewers surveyed relative to viewing time matches the frequency distribution of the criteria. A plurality of television devices each comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channels or broadcasting station networks were viewed via the receiver and at what time, wherein for each of the plurality of television devices belonging to a predetermined survey area, when the viewing time of television broadcasts within a predetermined unit time interval is given for each attribute included in the viewer attribute configuration of each television device, a first process identifies the frequency distribution of the number of viewers with respect to viewing time for each attribute in the survey area based on said viewing time,
The system is configured to perform a second process on the computer, which involves comparing the frequency distribution of the number of viewers for each attribute identified by the first process with the frequency distribution of predetermined criteria for each attribute in the survey area, and then selecting viewers to be surveyed for each attribute in the survey area.
The second process is a viewing information processing program characterized in that it causes the computer to select viewers to be surveyed for each attribute such that the frequency distribution of the number of viewers to be surveyed relative to viewing time matches the frequency distribution of the criteria.