JP7772726B2 - Viewing information processing device and viewing information processing method - Google Patents

Description

The present invention relates to a viewing information processing device and a viewing information processing method that have the function of estimating viewer attribute composition from viewing log information obtained for each television device.

In recent years, television devices equipped with the ability to transmit viewing data, including viewing log information indicating which channels were viewed and when, to television manufacturers, etc. via external networks such as the Internet have become widespread. In this specification, a device that can receive and view television broadcasts and has the ability to transmit viewing log information as described above is referred to as a television device.

As seen in Patent Document 1, for example, the present application proposes technology that uses a pre-trained model (mathematical model) to estimate the attribute composition of viewers of each television device (more specifically, one or more viewers who can watch television broadcasts received via that television device) from viewing log information obtained from that television device. Here, the attribute composition of viewers of a television device indicates which of multiple attribute categories, such as gender and age, the viewers are made up of.

Patent No. 6433615

As seen in Patent Document 1 above, when estimating the attributes of the audience members from viewing log information obtained from each television device, the estimated viewer attribute composition may be an abnormal attribute composition that is likely to be incorrect. For example, if the estimated viewer attributes are only young children, or if the number of estimated viewer attribute types is zero (in other words, if it is estimated that there are no viewers of the television device), or if the number of estimated viewer attribute types is too large, the estimated viewer attribute composition may be considered to be incorrect or an abnormal attribute composition that is likely to be incorrect.

Furthermore, if viewing log information from a television device with an estimated abnormal attribute composition and the attribute composition of the viewer are used to investigate viewing conditions such as viewer ratings by attribute, there is a risk that the reliability of the investigation results will be compromised.

Here, it is possible to exclude viewing log information for television devices with estimated abnormal attribute configurations from the target of use in the viewing status survey, but in that case, there is a risk that the total number of target television devices may not be sufficient.

The present invention was made in light of this background, and aims to provide a viewing information processing device or viewing information processing method that has the function of estimating the attribute composition of viewers of a television device from the viewing log information of the television device, and that, if an abnormal attribute composition is estimated, can appropriately correct the abnormal attribute composition.

In order to achieve the above object, the viewing information processing device of the present invention includes a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver, and a viewing log information acquisition unit that acquires viewing log information output from the television device;
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device ;
The attribute composition estimation unit is configured to calculate the sum of the values of the attribute-specific presence probability, select a predetermined number of people based on the sum in descending order of the attribute-specific presence probability values, and estimate the number of people as the attributes of the constituent members of the viewers of the television device (first invention).

According to the first aspect of the present invention, if the attribute configuration estimated by the attribute configuration estimation unit for a viewer of a television device is an abnormal attribute configuration, the attribute configuration of the viewer of the television device is corrected in accordance with the attribute-specific existence probability for each attribute included in the attribute configuration. This allows the attribute configuration to be corrected in accordance with the attribute-specific existence probability based on actual viewing log information for the television device. This makes it possible to appropriately correct the attribute configuration of the viewer of the television device from the estimated abnormal attribute configuration.

Furthermore, according to the first invention , the attribute composition of the viewer of the television device can be easily and appropriately corrected based on the sum of the values of the attribute-specific existence probability for each attribute included in the attribute composition estimated by the attribute composition estimation unit for the viewer of the television device.
The attribute composition estimation unit may be configured to estimate an attribute for which the value of the attribute-specific existence probability is equal to or greater than a predetermined threshold as an attribute of the constituent members of the viewers of the television device.

This allows the attribute composition of viewers of a television device to be easily and appropriately corrected based on the attribute-specific presence probability values for each attribute included in the attribute composition estimated by the attribute composition estimation unit for viewers of that television device.

In the present invention, with respect to any numerical value such as age or probability, "greater than or equal to A" may mean either greater than A, or greater than A or equal to A. Similarly, "equal to or less than A" may mean either smaller than A, or smaller than A or equal to A.

The viewing information processing device of the present invention includes a television receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver. The television receiver includes a viewing log information acquisition unit that acquires viewing log information output from the television device.
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a first attribute configuration in which an attribute configuration of a viewer of the television device consists solely of a child attribute, the child attribute being an attribute of a child under a first predetermined age;
When the attribute configuration estimated by the attribute configuration estimation unit is the first attribute configuration, the attribute configuration correction unit corrects the attribute configuration of the viewer of the television device to an attribute configuration that includes the child attribute and attribute A, which is the attribute identified by the attribute configuration estimation unit and has the highest value of the attribute-specific existence probability among attributes other than the child attribute that satisfy the condition of being above a second predetermined age ( second invention).

This allows the attribute configuration estimated as the first attribute configuration to be corrected to include not only child attributes, but also attribute A, which is the attribute other than child attributes that meet the condition of being above a second predetermined age and is most likely to represent a television device viewer, i.e., to have an attribute configuration similar to that of a typical household with children. This allows the attribute configuration of television device viewers to be appropriately corrected from the first attribute configuration.

In the second invention, the attribute composition correction unit may be configured to correct the attribute composition of the viewer of the television device so as to further include, when the attribute composition estimated by the attribute composition estimation unit is the first attribute composition, an attribute of an age between the child attribute and an attribute that satisfies the condition of being above the second specified age, if the value of the attribute-specific existence probability identified by the attribute composition estimation unit is the same as or higher than Pa, which is the value of the attribute-specific existence probability for attribute A ( third invention).

As a result, even if an age attribute falls between the child attribute and an attribute that satisfies the condition of being above the second predetermined age, if the attribute-specific presence probability value is high (if there is a high possibility that the person is a television device viewer), that attribute can be further included in the attribute configuration of television device viewers.

The viewing information processing device of the present invention includes a television receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver. The television receiver includes a viewing log information acquisition unit that acquires viewing log information output from the television device.
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a second attribute configuration in which the number of types of attributes included in the attribute configuration of the viewer of the television device is zero;
When the attribute configuration estimated by the attribute configuration estimation unit is the second attribute configuration, the attribute configuration correction unit corrects the attribute configuration of the viewer of the television device to an attribute configuration that includes attribute B, which is the attribute with the highest value of the attribute-specific existence probability identified by the attribute configuration estimation unit, among the attributes that satisfy the condition of being a third predetermined age or older ( fourth invention).

When the fourth invention is combined with the second or third invention, the attribute that satisfies the condition of being equal to or greater than the third predetermined age is an attribute of an age higher than the child attribute of being equal to or less than the first predetermined age. The third predetermined age may be lower than the second predetermined age, the same as the second predetermined age, or higher than the second predetermined age.

According to the fourth aspect of the present invention, the attribute configuration estimated as the second attribute configuration can be corrected to include, among the attributes that satisfy the condition of being a third predetermined age or older, the attribute configuration that is most likely to be of a television device viewer. Thus, the attribute configuration of television device viewers can be appropriately corrected from the second attribute configuration.

In the above-mentioned fourth invention, the attribute that satisfies the condition of being above the third predetermined age is an attribute other than a child attribute, which is an attribute of a child below a first predetermined age, and the attribute composition correction unit may be configured to correct the attribute composition of the viewer of the television device to further include the child attribute when the attribute composition estimated by the attribute composition estimation unit is the second attribute composition, and the value of the attribute-specific presence probability identified by the attribute composition estimation unit for the child attribute is the same as or higher than Pb, which is the value of the attribute-specific presence probability identified for attribute B, and when attribute B is an attribute that satisfies the condition of being above a fourth predetermined age that is higher than the third predetermined age ( fifth invention).

When the fifth invention is combined with the second or third invention, the attribute that satisfies the condition of being equal to or greater than the fourth predetermined age is an attribute of an age higher than the child attribute of being equal to or less than the first predetermined age. The fourth predetermined age may be lower than the second predetermined age, the same as the second predetermined age, or higher than the second predetermined age.

According to the fifth invention, when the value of the attribute-specific existence probability identified for a child attribute is high (when there is a high possibility that the person is a viewer of the television device), and when attribute B is an attribute that satisfies the condition of being above the fourth predetermined age, the child attribute can be further included in the attribute configuration of the viewer of the television device.

In the first to fifth inventions, the predetermined abnormal attribute configuration may include a third attribute configuration in which the number of types of attributes of the viewer of the television device is equal to or greater than a predetermined number. In this case, when the attribute configuration estimated by the attribute configuration estimation unit is the third attribute configuration, the attribute configuration correction unit may correct the attribute configuration of the viewer of the television device to an attribute configuration in which one or more attributes, including at least the attribute with the lowest attribute-specific existence probability value identified by the attribute configuration estimation unit, are deleted from the attributes included in the third attribute configuration (a sixth invention).

According to the sixth aspect of the present invention, the attribute configuration estimated as the third attribute configuration can be corrected by deleting one or more attributes from the third attribute configuration, the attribute including at least an attribute that is least likely to be associated with a television device viewer. This makes it possible to appropriately correct the attribute configuration of television device viewers from the third attribute configuration.

In order to achieve the above object, the viewing information processing method of the present invention includes a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
The second step is characterized in that it is configured to calculate the sum of the values of the attribute-specific presence probability, select a predetermined number of people based on the sum in order from the highest value of the attribute-specific presence probability, and estimate this as the attribute of the constituent members of the viewers of the television device (seventh invention).
The viewing information processing method of the present invention includes a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a first attribute configuration in which an attribute configuration of a viewer of the television device consists solely of a child attribute, the child attribute being an attribute of a child under a first predetermined age;
The third step is characterized in that, when the attribute configuration estimated in the second step is the first attribute configuration, the attribute configuration of the viewer of the television device is corrected to an attribute configuration that includes the child attribute and attribute A, which is the attribute identified in the second step and has the highest value of the attribute-specific existence probability among attributes other than the child attribute that satisfy the condition of being above a second predetermined age (8th invention).
The viewing information processing method of the present invention includes a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a second attribute configuration in which the number of types of attributes included in the attribute configuration of the viewer of the television device is zero;
The third step is characterized in that, when the attribute configuration estimated in the second step is the second attribute configuration, the attribute configuration of the viewers of the television device is corrected to an attribute configuration that includes attribute B, which is the attribute with the highest value of the attribute-specific existence probability identified in the second step, among the attributes that satisfy the condition of being a third predetermined age or older (ninth invention).

According to the seventh to ninth inventions, if the attribute configuration estimated by the second step for a viewer of a television device is an abnormal attribute configuration, the attribute configuration can be corrected according to the attribute-specific existence probability, as in the first , second, and fourth inventions , respectively , so that the attribute configuration can be appropriately corrected from the abnormal attribute configuration.

1 is a diagram showing an overall system according to an embodiment of the present invention. 2 is a block diagram for explaining the processing of an attribute configuration estimation unit of the viewing information processing device shown in FIG. 1 . 4 is a flowchart for explaining the processing of an attribute configuration correction unit of the viewing information processing device shown in FIG. 1 . 2 is a block diagram for explaining the processing of an attribute-specific viewing information generating unit of the viewing information processing device shown in FIG. 1 ;

One embodiment of the present invention will be described below with reference to Figures 1 to 4. In the following description of this embodiment, the terms "greater than or equal to," "less than or equal to," and "less than" will be used in their usual sense. That is, when A is greater than or equal to B, it means A≧B; when A is less than or equal to B, it means A≦B; and when A is less than B, it means A<B.

Referring to FIG. 1, the system described in this embodiment has a viewing information processing device 1 that executes processing related to surveys of television broadcast viewing conditions. The viewing information processing device 1 is composed of, for example, one or more computers. The computer includes a processor such as a CPU, memory (storage device), an interface circuit, a communication device, etc. (not shown). The viewing information processing device 1 can acquire viewing data related to television broadcast viewing at each device-specific household 20 from multiple device-specific households 20 belonging to a survey area for viewing conditions, etc. (for example, a prefecture-based region, or a region combining multiple prefectures such as the Kanto region or Kinki region). The viewing data includes viewing log information indicating which channel (or which broadcasting station affiliate) the television broadcast was viewed on and when.

Here, in this embodiment, each "device-specific household 20" refers to a household that includes a television device 23 that includes a television broadcast receiver 21 and a viewing data output device 22 that can output viewing data related to the viewing of the television broadcast, and that includes one or more viewers of the television device 23 (more specifically, one or more viewers who can watch the television broadcast received by the receiver 21 of the television device 23) as its members. In other words, a device-specific household 20 refers to a combination of a television device 23 and a viewer of the television device 23.

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 the 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.

The television device 23 of each device-specific household 20 may be composed of a television, or may be composed of a television and a recording device connected to it. The viewing data output device 22 of the television device 23 is composed of, for example, a processor such as a microcomputer (not shown), memory, an interface circuit, a communication device, etc. The viewing data output device 22 is capable of detecting the channel of the television broadcast viewed via the television device 23 that includes it, as well as the date and time when the television broadcast on that channel was viewed (more specifically, the date and time when viewing of the television broadcast on that channel began and ended). From this detected information, viewing log information for the device-specific household 20 (hereinafter referred to as device-specific viewing log information) can be generated. In other words, the device-specific viewing log information is the overall viewing log information of the members included in the device-specific household 20.

The viewing data output device 22 can then periodically (or upon request from) 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 that constitutes the television device 23 via an external network NW, such as the Internet or a telephone line network. In this case, the viewing data transmitted from the viewing data output device 22 includes not only device-specific viewing log information, but also location information (e.g., information indicating the first three digits of the postal code) that has been pre-registered on the television or recording device that constitutes the television device 23, and identification information for the television device 23. Note that the identification information for 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 includes a viewing data acquisition unit 11, an attribute configuration estimation unit 12, an attribute configuration correction unit 13, an attribute-specific viewing information generation unit 14, and a viewing situation estimation unit 15, as functions realized by the implemented hardware configuration and program (software configuration). Below, we will explain the overall processing of the viewing information processing device 1, along with details of each of these functional units.

The viewing data acquisition unit 11 is capable of communicating with a server (not shown) of the manufacturer of each television device 23 in the device-specific household 20, and by communicating in this manner, is able to acquire viewing data for each television device 23 of that manufacturer. The viewing data acquisition unit 11 functions as a viewing log information acquisition unit in the present invention.

In addition, if the viewing information processing device 1 is able to communicate with the viewing data output device 22 of the device-specific household 20, the viewing data acquisition unit 11 may acquire viewing data of the television device 23 of the device-specific household 20 directly from the viewing data output device 22 of that television device 23. In addition, the viewing information processing device 1 may acquire viewing data of the television device 23 from each manufacturer via an appropriate storage device.

The viewing information processing device 1 sequentially executes the processing of the attribute composition estimation unit 12, attribute composition correction unit 13, attribute-specific viewing information generation unit 14, and viewing situation estimation unit 15 for each device-specific household 20 in the survey area using viewing data acquired by the viewing data acquisition unit 11.

The attribute composition estimation unit 12 is a functional unit that estimates the attribute composition indicating which members of which attributes are included in the device-specific household 20 (in other words, the attribute composition of viewers of the television devices 23 of the 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 devices 23 of each device-specific household 20.

Here, the attributes of the members (viewers of television devices 23) of each device-specific household 20 are classified into multiple types of attributes, for example, according to the age and gender of the members. For example, they are classified into multiple types of attributes such as children under y1, men aged y1 or older but under y2, women aged y1 or older but under y2, men aged y2 or older, and women aged y2 or older. Hereinafter, the number of types of attributes will be designated as N, and each of the N types of attributes will be appropriately represented as At(i) (i = 1, 2, ..., N). Note that the attributes of the members of each device-specific household 20 can be classified according to various parameters other than age and gender, such as occupation and educational background.

As shown in FIG. 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 previously subjected to machine learning processing. Here, the first model is a model that has been previously subjected to machine learning processing so that the attribute composition of the members of any device-specific household 20 belonging to the survey area can be estimated from the device-specific viewing logs for a predetermined period of that device-specific household 20. In the machine learning processing for the first model, viewing log information for each sample household obtained from people meters or the like installed in each of multiple sample households whose attribute composition is known, and the attribute composition for each sample household, are used as learning data. Furthermore, a publicly known algorithm can be used as the algorithm for the machine learning processing.

In this case, in this embodiment, the first model is configured to be able to determine, for each of N types of attributes At(i) (i = 1, 2, ..., N), an attribute-specific presence probability, which is the probability that a member of each attribute At(i) is present in the device-specific household 20, from the 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 may be adopted. However, the first model may be of another form, as long as it is able to determine an attribute-specific presence probability (or an index value similar thereto) from the device-specific viewing log information for a predetermined period of time for the device-specific household 20.

The attribute composition estimation unit 12 then estimates attributes for which the attribute-specific presence probability identified by the first model is equal to or greater than a predetermined threshold (e.g., 0.5 or greater) as attributes of the members of the device-specific household 20. As an example, Figure 2 illustrates the processing of the attribute composition estimation unit 12 for one device-specific household 20 (in Figure 2, the device-specific household 20 whose identification information ID is x1). Note that in Figure 2 and the block diagram shown in Figure 4 described below, the processing execution unit is indicated by a thick frame, and the data output unit or data is indicated by a thin frame.

In the example shown in Figure 2, the attribute-specific presence probabilities of members of a device-specific household 20 with identification information ID x1 for a predetermined period are identified by the first model as shown in the figure, based on device-specific viewing log information. In this example, of the N types of attributes At(i) (i = 1, 2, ..., N), the attribute-specific presence probabilities of each of attributes At(1), At(3), and At(N) are greater than or equal to a predetermined threshold (here, for example, 0.5), and the attribute-specific presence probabilities of each of the other attributes At(2), At(4) to At(N-1) are less than the threshold.

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

In this way, the attribute composition estimation unit 12 estimates whether or not members (viewers) of each attribute are present in the device-specific household 20 based on whether or not the attribute-specific presence probability identified by the first model for each device-specific household 20 is equal to or greater than a predetermined threshold, thereby estimating the attribute composition of the members of the device-specific household 20.

After executing the processing of the attribute composition estimation unit 12, the viewing information processing device 1 then executes the processing of the attribute composition correction unit 13. This attribute composition correction unit 13 is a functional unit that corrects the estimated result of the attribute composition of the members of each device-specific household 20 when the attribute composition estimated by the attribute composition estimation unit 12 for that device-specific household 20 is a predetermined abnormal attribute composition that is predefined as an attribute composition that is likely to be incorrect.

In this embodiment, for example, three types of attribute configurations (first to third attribute configurations) are predefined as predetermined abnormal attribute configurations. The first attribute configuration is an attribute configuration in which the members estimated by the attribute configuration estimation unit 12 for the device-specific household 20 consist only of members with the attribute of children under the age of y1 (or y1 or less) (hereinafter, this attribute will be referred to as the At(1) attribute) among the N types of attributes At(i) (i = 1, 2, ..., N). In other words, this first attribute configuration is an attribute configuration in which, among the attribute-specific presence probabilities identified for each of the N types of attributes At(i) (i = 1, 2, ..., N), only the attribute-specific presence probability corresponding to the attribute At(1) of children under the age of y1 (or y1 or less) (hereinafter, simply referred to as the child attribute At(1)) is equal to or exceeds a predetermined threshold. The above-mentioned y1 years old corresponds to the first predetermined age in this invention.

The second attribute configuration is an attribute configuration in which the attribute configuration estimated by the attribute configuration estimation unit 12 for the device-specific household 20 does not include any of the N types of attributes At(i) (i = 1, 2, ..., N) (i.e., the total number of attributes included in the attribute configuration is zero). In other words, the second attribute configuration is an attribute configuration in which all of the attribute-specific existence probabilities identified for each of the N types of attributes At(i) (i = 1, 2, ..., N) are smaller than a predetermined threshold value.

The third attribute configuration is an attribute configuration in which the attribute configuration estimated by the attribute configuration estimation unit 12 for the device-specific household 20 has a predetermined number (e.g., 5 or 6) or more types of attributes. In other words, the third attribute configuration is an attribute configuration in which, of the attribute-specific existence probabilities identified for each of N types of attributes At(i) (i = 1, 2, ..., N), the attribute-specific existence probability corresponding to each of the predetermined number or more types of attributes is equal to or greater than a predetermined threshold. In the following description, the predetermined number related to the third attribute configuration will be referred to as the predetermined number nx.

These first to third attribute configurations are either attribute configurations that can be considered not to exist in reality (for example, the first attribute configuration or the second attribute configuration), or attribute configurations that can be considered to exist less frequently than attribute configurations other than the first to third attribute configurations (for example, the third attribute configuration).

The attribute configuration correction unit 13 receives the attribute configuration estimated by the attribute configuration estimation unit 12 for each device-specific household 20 and the attribute-specific presence probability identified by the first model of the attribute configuration estimation unit 12. The attribute configuration correction unit 13 then determines, for each device-specific household 20, whether the estimated attribute configuration is an abnormal attribute configuration of any of the first to third attribute configurations described above, and corrects the attribute configuration if the estimated attribute configuration is an abnormal attribute configuration.

Specifically, the attribute configuration correction unit 13 executes the processing shown in the flowchart of Figure 3 for each device-specific household 20, thereby correcting the attribute configuration for device-specific households 20 for which the attribute configuration estimated by the attribute configuration estimation unit 12 is an abnormal attribute configuration.

In STEP 1, the attribute configuration correction unit 13 determines whether the attribute configuration estimated for each device-specific household 20 is the third attribute configuration (an attribute configuration including a predetermined number nx or more types of attributes). If the result of this determination is positive (if the estimated attribute configuration is the third attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP 2.

In STEP 2, the attribute configuration correction unit 13 corrects the attribute configuration of the device-specific household 20 (the device-specific household 20 for which the judgment result in STEP 1 was positive) to remove attributes selected in accordance with the attribute-specific presence probability identified for the device-specific household 20, thereby reducing the total number of attributes included in the attribute configuration of the device-specific household 20 to a total number less than the predetermined number nx.

Specifically, the attribute configuration correction unit 13 selects one or more types of attributes with lower probability values (probability values closer to the lowest probability value (including the lowest probability value) than the highest probability value) from the attribute-specific existence probabilities corresponding to each of the multiple attributes (a predetermined number nx or more types of attributes) included in the attribute configuration estimated for the device-specific household 20, and corrects the attribute configuration by deleting the selected attributes to be deleted from the attribute configuration of the device-specific household so that the total number of attributes after deletion is less than the predetermined number nx. In this case, of the multiple attributes included in the estimated attribute configuration, attributes with lower attribute-specific existence probability values (probability values) are preferentially deleted.

The process of reducing the total number of attributes included in the attribute configuration as described above can be performed, for example, as follows. That is, the attribute configuration correction unit 13 repeatedly selects and deletes attributes (attributes with lower probability values) with relatively low attribute-specific existence probability values (probability values) from among the multiple attributes included in the estimated attribute configuration (attributes of a predetermined number nx or more types), starting with the attribute with the lowest probability value, until the total number of attributes after deletion falls below the predetermined number nx. In this case, if, at each iteration, there are multiple attributes with the lowest attribute-specific existence probability values among the attributes included in the attribute configuration before deletion, these multiple attributes are deleted.

As an example, assume that the predetermined number nx is, for example, "5," the attribute composition estimated by the attribute composition estimation unit 12 for a certain device-specific household 20 is an attribute composition consisting of seven types of attributes, At(1) to At(7), and the attribute-specific presence probability values for At(1) to At(7) are 0.63 (: At(1)), 0.53 (: At(2)), 0.53 (: At(3)), 0.70 (: At(4)), 0.80 (: At(5)), 0.57 (: At(6)), and 0.57 (: At(7)).

In this case, from the attributes At(1) to At(7), the two attributes At(2) and At(3) with the lowest attribute-specific presence probability values are deleted, and then the two attributes At(6) and At(7) with the next lowest attribute-specific presence probability values are deleted. As a result, the attribute configuration for the device-specific household 20 is corrected to an attribute configuration consisting of the three attributes At(1), At(4), and At(5).

In this embodiment, the processing of STEP 2 (processing to correct the attribute configuration of the device-specific household 20 from the third attribute configuration) is performed as described above. As a result, the attribute configuration of the device-specific household 20 whose estimated attribute configuration is the third attribute configuration is corrected to an attribute configuration in which the number of attribute types included therein is less than a predetermined number nx.

If the determination result of STEP 1 is negative (if the estimated attribute configuration is not the third attribute configuration), the attribute configuration correction unit 13 executes the process of STEP 3. The attribute configuration correction unit 13 also executes the process of STEP 3 after executing the process of STEP 2. In STEP 3, the attribute configuration correction unit 13 determines whether the estimated attribute configuration is the first attribute configuration (an attribute configuration including only the child attribute At(1)).

In this embodiment, the attribute configuration corrected by the processing of STEP 2 may become the first attribute configuration. For example, if the attribute configuration before correction by the processing of STEP 2 (= the third attribute configuration) includes the child attribute At(1), and the attribute-specific existence probability values of other attributes in the attribute configuration before correction are lower than the attribute-specific existence probability value of the child attribute At(1) and are the same, then the attribute configuration after correction by the processing of STEP 2 will become an attribute configuration that includes only the child attribute At(1), i.e., the first attribute configuration. For this reason, in this embodiment, the attribute configuration correction unit 13 performs the determination processing of STEP 3 not only when the determination result of STEP 1 is negative, but also after the processing of STEP 2 has been performed.

If the determination result in STEP 3 is positive (if the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration after correction in STEP 2 is the first attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP 4.

In STEP 4, the attribute configuration correction unit 13 corrects the attribute configuration of the device-specific household 20 (the device-specific household 20 for which the judgment result in STEP 3 was positive) so that the attribute configuration includes, in addition to the child attribute At(1), one or more attributes selected according to the attribute-specific presence probability identified for the device-specific household 20.

Specifically, the attribute configuration correction unit 13 selects, as an attribute to be added, the attribute with the highest attribute-specific presence probability from among the N types of attributes At(i) (i = 1, 2, ..., N) for which the attribute-specific presence probability has been determined for the device-specific household 20 by the attribute configuration estimation unit 12, which meets the condition of being a predetermined age or older (adult age), and corrects the attribute configuration of the device-specific household 20 so that the selected attribute to be added is added to the attribute configuration of the device-specific household 20 (an attribute configuration consisting only of the child attribute At(1)). In this case, if there are multiple attributes with the highest attribute-specific presence probability, these multiple attributes are added to the attribute configuration of the device-specific household 20. Note that the above-mentioned predetermined age corresponds to the second predetermined age or fourth predetermined age in this invention.

For example, assume that the attribute configuration estimated by the attribute configuration estimation unit 12 for a certain device-specific household 20 is the first attribute configuration, the attributes that satisfy the condition of being above a certain age (adult age) are four types of attributes At(4) to At(7), and the attribute-specific presence probability values for At(4) to At(7) are 0.45 (: At(4)), 0.36 (: At(5)), 0.45 (: At(6)), and 0.28 (: At(7)). In this case, of At(4) to At(7), the two types of attributes At(4) and At(6) have the highest attribute-specific presence probability values, so these attributes At(4) and At(6) are selected as attributes to be added and added to the attribute configuration of the device-specific household 20. As a result, the attribute configuration of the device-specific household 20 is corrected from the first attribute configuration to an attribute configuration including three types of attributes: At(1), At(4), and At(6).

The same applies to a device-specific household 20 whose attribute configuration after correction in STEP 2 has become the first attribute configuration. Additionally, in the processing of STEP 4, the attribute configuration after correction may become an attribute configuration (= third attribute configuration) that includes more than the predetermined number nx of types of attributes. Note that in the processing of STEP 4, the attribute configuration of the device-specific household 20 may be corrected so that it can include not only an attribute that satisfies the condition of being above a predetermined age (adult age) (hereinafter referred to as the adult attribute), but also an attribute of an age between the adult attribute and the child attribute At(1) (hereinafter referred to as the young person attribute).

Specifically, if the attribute-specific presence probability value for a certain young person attribute is equal to or greater than the highest attribute-specific presence probability value Pa among the attribute-specific presence probabilities of the adult attributes (either the same value as Pa or a value higher than Pa), not only the attribute whose attribute-specific presence probability value is Pa, but also young person attributes whose attribute-specific presence probabilities are equal to or greater than Pa may be added to the attribute configuration of the device-specific household 20.

For example, let us assume that there are four types of adult attributes, At(4) to At(7), and two types of youth attributes, At(2) and At(3), and the values of the attribute-specific presence probability for At(2) to At(7) in a device-specific household 20 (a device-specific household 20 whose attribute configuration is estimated to be the first attribute configuration) are 0.49 (: At(2)), 0.30 (: At(3), 0.45 (: At(4)), 0.36 (: At(5)), 0.45 (: At(6)), and 0.28 (: At(7)). In this case, Of the person attributes At(4) to At(7), attributes At(4) and At(6) with the highest attribute-specific presence probability values, and of the youth attributes At(2) and At(3), attribute At(2) with a higher attribute-specific presence probability value than attributes At(4) and At(6) are selected as attributes to be added and added to the attribute configuration of the device-specific household 20. As a result, the attribute configuration of the device-specific household 20 is corrected from the first attribute configuration to an attribute configuration including four types of attributes At(1), At(2), At(4), and At(6).

In this embodiment, the process of STEP 4 (processing of correcting the attribute configuration of the device-specific household 20 from the first attribute configuration) is performed as described above.
As a result, the attribute configuration of the device-specific household 20 whose attribute configuration estimated by the attribute configuration estimation unit 12 or whose attribute configuration after correction in STEP 2 has become the first attribute configuration is corrected to become an attribute configuration that further includes attributes other than At(1) (adult attributes). Note that when the processing of STEP 4 is executed, the processing of the attribute configuration correction unit 13 for the device-specific household 20 whose attribute configuration has been corrected by that processing is terminated (and thus the attribute configuration of the device-specific household 20 is confirmed).

If the determination result in STEP 3 is negative (if the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration after correction in STEP 2 is not the first attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP 5. In STEP 5, the attribute configuration correction unit 13 determines whether the estimated attribute configuration is the second attribute configuration (an attribute configuration in which the total number of included attributes is zero).

In this embodiment, the attribute configuration corrected by the processing of STEP 2 may become the second attribute configuration. For example, if the attribute-specific existence probability values of multiple attributes (attributes with a number of types equal to or greater than the predetermined number nx) included in the attribute configuration (= third attribute configuration) before correction by the processing of STEP 2 are all the same, the attribute configuration corrected by the processing of STEP 2 will be an attribute configuration in which the total number of attributes included therein is zero, i.e., the second attribute configuration. For this reason, in this embodiment, the attribute configuration correction unit 13 performs the determination processing of STEP 5 not only when the determination results of both STEPs 1 and 3 are negative, but also when the determination result of STEP 3 becomes negative after the processing of STEP 2 is performed.

If the determination result in STEP 5 is positive (if the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration corrected in STEP 2 is the second attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP 6.

In this STEP 6, the attribute configuration correction unit 13 corrects the attribute configuration of the device-specific household 20 (the device-specific household 20 for which the judgment result in STEP 5 was positive) so that the attribute configuration includes one or more attributes selected according to the attribute-specific presence probability identified for the device-specific household 20.

Specifically, the attribute composition correction unit 13 selects, as an attribute to be added, the attribute with the highest attribute-specific presence probability from the attributes At(2) to At(N), excluding the child attribute At(1), among the N types of attributes At(i) (i = 1, 2, ..., N) for which the attribute composition estimation unit 12 has determined the attribute-specific presence probability for the device-specific household 20, and corrects the attribute composition of the device-specific household 20 so that the attribute composition consisting of the selected attributes to be added becomes the attribute composition of the device-specific household 20. In other words, the attributes At(2) to At(N), excluding the child attribute At(1), are attributes that satisfy the condition of being equal to or greater than (or higher than) the predetermined age y1 associated with the child attribute At(1), and the predetermined age y1 corresponds to the third predetermined age in the present invention.

In this case, if there are multiple attributes among the attributes At(2) to At(N), excluding the child attribute At(1), that have the highest attribute-specific existence probability, the attribute configuration of the device-specific household 20 is corrected so that it is composed of these multiple attributes.

For example, suppose the attribute configuration estimated by the attribute configuration estimation unit 12 for a certain device-specific household 20 is the second attribute configuration, and of the attributes At(2) to At(N), the attribute-specific existence probability values for At(3) and At(4) are both 0.47, and the attribute-specific existence probability values for the other attributes At(2), At(5) to At(N) are all lower than 0.47. In this case, of the attributes At(2) to At(N), the two attributes At(3) and At(4) have the highest attribute-specific existence probability values. Therefore, these attributes At(3) and At(4) are selected as attributes to be added, and the attribute configuration of the device-specific household 20 is corrected to an attribute configuration consisting of these attributes At(3) and At(4) to be added.

The same applies to a device-specific household 20 whose attribute configuration after correction in STEP 2 has become the second attribute configuration. Additionally, in the processing of STEP 6, the attribute configuration after correction may become an attribute configuration (= third attribute configuration) that includes more than the predetermined number nx of attributes. Note that in the processing of STEP 6, the attribute configuration of the device-specific household 20 may be corrected so that it can include not only attributes other than the child attribute At(1), but also the child attribute At(1).

Specifically, if the attribute-specific presence probability value for the child attribute At(1) is equal to or greater than the highest attribute-specific presence probability value Pb among the attribute-specific presence probabilities other than the child attribute At(1) (either the same value as Pb or a value higher than Pb), and the attribute with the attribute-specific presence probability value Pb includes an adult attribute, then the child attribute At(1) may be added to the attribute configuration of the device-specific household 20 in addition to the attribute with the attribute-specific presence probability value Pa.

For example, assume that there are four types of adult attributes, At(4) to At(7), and two types of youth attributes, At(2) and At(3), and the attribute-specific existence probability of the child attribute At(1) in device-specific household 20 (device-specific household 20 whose attribute configuration is estimated to be the second attribute configuration) is 0.48, and that of the attributes At(2) to At(7), the attribute-specific existence probability values of the youth attribute At(3) and the adult attribute At(4) are both 0.47, and that the attribute-specific existence probability values of the other attributes At(2) and At(5) to At(7) are all lower than 0.47. Of the attributes At(2) to At(7) other than the child attribute At(1), the attributes At(3) and At(4) with the highest attribute-specific presence probability values include the adult attribute At(4), and the attribute-specific presence probability value of the child attribute At(1) is higher than the attribute-specific presence probability values of the attributes At(3) and At(4). Therefore, not only the attributes At(3) and At(4), but also the child attribute At(1) are selected as attributes to be added, and the attribute configuration of the device-specific household 20 is corrected to an attribute configuration consisting of these attributes At(1), At(3), and At(4) to be added.

Note that if the only attribute with the highest attribute-specific existence probability value among the attributes At(2) to At(7) is the young person attribute (for example, At(3)), then even if the child attribute At(1) has an attribute-specific existence probability value equal to or greater than the attribute-specific existence probability value of At(3), the child attribute At(1) will not be selected as an attribute to be added. This prevents the attribute configuration after correction from consisting solely of the child attribute At(1) and the young person attribute.

In this embodiment, the processing of STEP 6 (processing to correct the attribute configuration of the device-specific household 20 from the second attribute configuration) is performed as described above. As a result, the attribute configuration of the device-specific household 20 whose second attribute configuration is the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration corrected in STEP 2 is corrected to include one or more attributes.

In this embodiment, the processing of the attribute configuration correction unit 13 is executed as described above. As a result, in a device-specific household 20 for which the attribute configuration estimated by the attribute configuration estimation unit 12 is the first attribute configuration, the attribute configuration of the device-specific household 20 can be corrected so that, in addition to the child attribute At(1), at least an attribute (adult attribute) that meets the condition of being an adult of a predetermined age or older is included. In this case, the attributes added other than the child attribute At(1) are attributes with a higher attribute-specific presence probability value than the other attributes, so it is possible to minimize the deviation of the corrected attribute configuration from the actual attribute configuration of the device-specific household 20.

Furthermore, in a device-specific household 20 in which the attribute configuration estimated by the attribute configuration estimation unit 12 is the second attribute configuration, the attribute configuration of the device-specific household 20 can be corrected to include one or more attributes other than the child attribute At(1). In this case, the attributes included in the corrected attribute configuration are attributes with a higher attribute-specific presence probability value than other attributes, so it is possible to minimize the deviation of the corrected attribute configuration from the actual attribute configuration of the device-specific household 20. Furthermore, because the attributes included in the corrected attribute configuration include at least attributes other than the child attribute At(1), it is possible to prevent the corrected attribute configuration from becoming an attribute configuration consisting only of the child At(1) attributes (= first attribute configuration).

Furthermore, in a device-specific household 20 in which the attribute configuration estimated by the attribute configuration estimation unit 12 is the third attribute configuration, the attribute configuration of the device-specific household 20 can basically be corrected so that the attribute configuration includes fewer types of attributes than the predetermined number nx. In this case, the attributes to be deleted from the attribute configuration before correction are attributes with a lower attribute-specific existence probability value than the other attributes, so it is possible to prevent the corrected attribute configuration from deviating as much as possible from the actual attribute configuration of the device-specific household 20.

Note that the attribute configuration after correction by the processing of STEP 4 or 6 may become a third attribute configuration that includes a predetermined number nx or more types of attributes, but the frequency of occurrence of such cases is generally low. Furthermore, the third attribute configuration is more likely to actually exist than the first or second attribute configuration. For this reason, in this embodiment, if the attribute configuration after correction by the processing of STEP 4 or 6 becomes a third attribute configuration, that attribute configuration is maintained.

The viewing information processing device 1 then executes processing by the attribute-specific viewing information generation unit 14. This attribute-specific viewing information generation unit 14 is a functional unit that generates attribute-specific viewing information indicating whether or not the television broadcast that is the subject of a viewing situation survey (hereinafter referred to as the surveyed broadcast) has been viewed, for each attribute included in the attribute configuration of each device-specific household 20 that has undergone processing by the attribute configuration correction unit 13 (a device-specific household 20 whose attribute configuration has been determined).

As shown in Figure 4, the attribute-specific viewing information generation unit 14 estimates attribute-specific viewing information for each attribute of the constituent factors of each device-specific household 20 from viewing log information for a predetermined period of time for each device-specific household 20 using a second model that has been subjected to machine learning processing in advance.

In this embodiment, the second model is a model created separately for each of N types of attributes At(i) (i = 1, 2, ..., N). The second model corresponding to each attribute At(i) is a model that has been subjected to machine learning processing in advance so that, when the channel (or broadcasting station network) of the survey target broadcast and date and time data indicating the date and time of the broadcast (month, day of the week, time slot, etc.) are specified, the probability that members (viewers) of the attribute At(i) have watched the survey target broadcast (hereinafter referred to as attribute-specific viewing probability) can be determined from the device-specific viewing logs for a specified period of the device-specific household 20 to which the attribute At(i) belongs.

Note that Figure 2 shows that the viewing log information for a device-specific household 20 having members with a certain attribute At(n) is used to determine an attribute-specific viewing probability, which is the probability that members with attribute At(n) watched a survey target broadcast specified by channel (or broadcast station network) and date and time data. The same applies to second models corresponding to other attributes.

In the machine learning process for the second model described above, the viewing log information for each member of each sample household, obtained from people meters or the like installed in each of multiple sample households with known attribute configurations, and the attribute configuration of each sample household, are used as learning data. A publicly known algorithm may be used as the algorithm for this machine learning process. Furthermore, such a second model may be similar to the mathematical model for determining individual viewing in Patent Document 1, for example. However, the second model may be of another form, as long as it can identify the attribute-specific viewing probability (or a similar index value) for the survey target broadcast for each attribute of the members of the device-specific household 20 from the device-specific viewing log information for the device-specific household 20 for a predetermined period of time.

The attribute-specific viewing information generation unit 14 then determines (estimates) for each attribute of each member of the device-specific household 20 whether or not the member of each attribute watched the survey target broadcast, based on the attribute-specific viewing probability identified by the second model and the device-specific viewing log information of the device-specific household 20 (device-specific viewing log information including log information on the broadcast date and time of the survey target broadcast).

Specifically, the attribute-specific viewing information generation unit 14 determines that the survey target broadcast was viewed by a member of the attribute if the attribute-specific viewing probability value (probability value) identified by the second model for the survey target broadcast for each attribute of the members of the device-specific household 20 is equal to or greater than a predetermined threshold (e.g., 0.5 or greater) and the viewing probability information of the device-specific household 20 confirms that the survey target broadcast was actually viewed on the television broadcast channel (or broadcast station affiliate) at the broadcast date and time of the survey target broadcast; if the condition is not met, it determines that the survey target broadcast was not viewed by a member of the attribute.

For example, Figure 4 illustrates a situation in which it is determined that the surveyed broadcast was viewed by a member with attribute At(n) when the attribute-specific viewing probability for attribute Ar(n) is equal to or greater than a predetermined threshold and when it is confirmed from device-specific viewing probability information that actual viewing occurred on the channel (or broadcast station affiliate) of the specified television broadcast on the date and time indicated by the date and time data of the television broadcast.

In this embodiment, the processing of the attribute-specific viewing information generation unit 14 is performed as described above. As a result, attribute-specific viewing information is generated for each attribute included in the attribute configuration determined by the attribute configuration correction unit 13 for each device-specific household 20, indicating whether or not the survey target broadcast, for which the channel (or broadcast station series) and date and time data are specified, was viewed.

The viewing information processing device 1 then executes processing by the viewing situation estimation unit 15. This viewing situation estimation unit 15 is a processing unit that estimates viewing situation data (data equivalent to audience ratings) for the survey target broadcast for each attribute of the overall members of the device-specific households 20 in the survey target area, based on the attribute-specific viewing information obtained by processing by the attribute-specific viewing information generation unit 14 for each device-specific household 20 in the survey target area.

In this case, in this embodiment, if the total number of constituent members for each attribute At(i) (i = 1, 2, ..., N) in all device-specific households 20 in the survey area is denoted as M(i), and the total number of constituent members who are determined to have watched the survey target broadcast for each attribute At(i) is denoted as m(i), the viewing situation estimation unit 15 calculates the viewing situation data of the survey target broadcast for each attribute At(i), for example, using the following formula (1).

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

This allows us to obtain viewing data equivalent to viewership rates by attribute in the surveyed area.

According to the viewing information processing device 1 of this embodiment described above, if the attribute configuration of each device-specific household 20 estimated by the attribute configuration estimation unit 12 from the device-specific viewing log information of each device-specific household 20 in the survey area is an abnormal attribute configuration of either the first attribute configuration, the second attribute configuration, or the third abnormal attribute, the attribute configuration is corrected according to the attribute-specific presence probability. This correction ensures that the survey area does not include device-specific households 20 with the first attribute configuration or the second attribute configuration, and that the number of device-specific households 20 with the third attribute configuration is kept to a sufficiently low number.

Therefore, the attribute composition of households by device in the survey area can be identified in a way that ensures high reliability. As a result, the reliability of the attribute-specific viewing information for households by device generated by the attribute-specific viewing information generation unit 14 and the reliability of the viewing situation data generated by the viewing situation estimation unit 15 can be improved. Ultimately, it is possible to properly survey the viewing situation in the survey area.

The present invention is not limited to the above-described embodiment, and other embodiments may be adopted. Some other embodiments are exemplified below.
In the above embodiment, the attribute configuration estimation unit 12 estimates an attribute whose attribute-specific presence probability specified by the first model is equal to or greater than a predetermined threshold (e.g., 0.5 or greater) as an attribute of the members of the device-specific household 20. However, this is not limited to this. Specifically, for a predetermined attribute determined based on the attribute-specific presence probability, the attribute configuration estimation unit 12 may calculate the sum of the attribute-specific presence probability values, select a predetermined number of people based on the sum in descending order of the attribute-specific presence probability values (e.g., if the sum is 1.8, which exceeds 1, then one person from the highest value; if the sum is 2.5, which exceeds 2, then two people from the highest values), and estimate the selected attribute as an attribute of the members of the device-specific household 20.
In this case, when there are multiple people with the same attribute-specific existence probability value, the target person to be selected as a priority may be selected randomly, but priorities may also be set based on information such as "family composition,""TVmanufacturer," and "viewing minutes."
In addition, a method may be adopted in which the estimated number of people is defined for each attribute (not limited to the total value of the attribute-specific existence probability), and the attribute of the constituent factor is determined in order of the highest existence probability value.
In addition, in the above embodiment, three types of attribute configurations, namely, the first attribute configuration, the second attribute configuration, and the third attribute configuration, are used as the anomalous attribute information, but, for example, only the first attribute configuration and the second attribute configuration may be used as the anomalous attribute configuration, and the third attribute configuration may be excluded from the anomalous attribute configuration. Alternatively, for example, whether or not to exclude the third attribute configuration may be selected for each survey target area, or may be selected depending on whether or not the proportion of device-specific households 20 with the third attribute configuration among all device-specific households 20 whose attribute configurations have been estimated by the attribute configuration estimation unit 12 is higher than a predetermined value.

Furthermore, in the above embodiment, when the estimated attribute configuration is a third attribute configuration and the attribute configuration includes multiple attributes that have the same attribute-specific existence probability value and that value is lower than the other attributes, all of the multiple attributes are selected as attributes to be deleted. However, in this case, for example, the attributes to be deleted may be selected randomly or according to a predetermined priority from the multiple attributes with the same attribute-specific existence probability value, as long as the total number of attributes included in the attribute configuration after correction (after one or more attributes are deleted) is less than the predetermined number nx.

In addition, in the above embodiment, when the estimated attribute configuration or the attribute configuration after correction from the third attribute configuration is the first attribute configuration, if the attribute configuration includes multiple attributes that satisfy the condition of being an attribute with an adult age and have the same attribute-specific existence probability values and that are higher than the other attributes, the attribute configuration is corrected so that all of those multiple attributes are added to the attribute configuration. However, in this case, for example, the attribute to be added may be selected randomly or according to a predetermined priority from the multiple attributes with the same attribute-specific existence probability values, within the range that satisfies the condition that the total number of attributes included in the attribute configuration after correction (after addition of one or more attributes) is less than the predetermined number nx. This also applies when a new attribute is added to the attribute configuration when the estimated attribute configuration or the attribute configuration after correction from the third attribute configuration is the second attribute.

Furthermore, in the above embodiment, the viewing information processing device 1 is equipped with an attribute-specific viewing information generation unit 14 and a viewing situation estimation unit 15, but the viewing information processing device of the present invention may not be equipped with an attribute-specific viewing information generation unit 14 or a viewing situation estimation unit 15. In that case, the attribute-specific viewing information generation unit 14 and the viewing situation estimation unit 15 may be equipped in a device separate from the viewing information processing device. Also, the attribute configuration estimated by the attribute configuration estimation unit 12 or corrected by the attribute configuration correction unit 13 may be used in processing separate from the attribute-specific viewing information generation unit 14 and the viewing situation estimation unit 15.

1... Viewing information processing device, 11... Viewing data acquisition unit (viewing log information acquisition unit), 12... Attribute composition estimation unit, 13... Attribute composition correction unit, 21... Receiver, 22... Viewing data output device, 23... Television device.

Claims (9)

a viewing log information acquisition unit that acquires viewing log information output from a television device that includes a television broadcast receiver and a viewing data output device that can output viewing log information indicating which channel or broadcast station television broadcast was viewed and when via the receiver;
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device ;
The attribute composition estimation unit is configured to calculate the sum of the values of the attribute-specific presence probability, select a predetermined number of people based on the sum in descending order of the attribute-specific presence probability, and estimate the number as an attribute of the constituent members of the viewers of the television device . a viewing log information acquisition unit that acquires viewing log information output from a television device that includes a television broadcast receiver and a viewing data output device that can output viewing log information indicating which channel or broadcast station television broadcast was viewed and when via the receiver;
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a first attribute configuration in which an attribute configuration of a viewer of the television device consists solely of a child attribute, the child attribute being an attribute of a child under a first predetermined age;
A viewing information processing device characterized in that, when the attribute configuration estimated by the attribute configuration estimation unit is the first attribute configuration, the attribute configuration correction unit corrects the attribute configuration of the viewer of the television device to an attribute configuration that includes the child attribute and attribute A, which is the attribute with the highest value of the attribute-specific existence probability identified by the attribute configuration estimation unit among attributes other than the child attribute that satisfy the condition of being above a second predetermined age . 3. The viewing information processing device according to claim 2 ,
a viewer information processing device configured to correct the attribute configuration of the viewer of the television device so that, when the attribute configuration estimated by the attribute configuration estimation unit is the first attribute configuration, if there is an attribute, excluding the child attribute and the attribute that satisfies the condition of being above the second predetermined age, whose attribute-specific existence probability value identified by the attribute configuration estimation unit is the same as or higher than Pa, which is the attribute-specific existence probability value for attribute A, to further include the present attribute . a viewing log information acquisition unit that acquires viewing log information output from a television device that includes a television broadcast receiver and a viewing data output device that can output viewing log information indicating which channel or broadcast station television broadcast was viewed and when via the receiver;
an attribute configuration estimation unit that estimates an attribute configuration of viewers of the television device from the viewing log information acquired by the viewing log information acquisition unit,
the attribute composition estimation unit is configured to, when estimating the attribute composition of viewers of the television device, specify, for each type of attribute, an attribute-specific existence probability, which is the probability that viewers having the attribute exist, and to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of the constituent members of the viewers of the television device;
an attribute configuration correction unit that, when the attribute configuration estimated by the attribute configuration estimation unit is a predetermined abnormal attribute configuration, determines an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a second attribute configuration in which the number of types of attributes included in the attribute configuration of the viewer of the television device is zero;
A viewing information processing device characterized in that, when the attribute configuration estimated by the attribute configuration estimation unit is the second attribute configuration, the attribute configuration correction unit corrects the attribute configuration of the viewer of the television device to an attribute configuration that includes attribute B, which is the attribute with the highest value of the attribute-specific existence probability identified by the attribute configuration estimation unit, among the attributes that satisfy the condition of being a third predetermined age or older . 5. The viewing information processing device according to claim 4,
the attribute that satisfies the condition of being equal to or greater than the third predetermined age is an attribute other than a child attribute, which is an attribute of a child equal to or less than the first predetermined age;
a viewer information processing device configured to correct the attribute composition of the viewer of the television device to further include the child attribute when the attribute composition estimated by the attribute composition estimation unit is the second attribute composition, the value of the attribute-specific presence probability identified by the attribute composition estimation unit for the child attribute is the same as or higher than Pb, which is the value of the attribute-specific presence probability identified for attribute B, and the attribute B is an attribute of a fourth predetermined age or above that is higher than the third predetermined age . The viewing information processing device according to any one of claims 1 to 5 ,
the predetermined abnormal attribute configuration includes a third attribute configuration in which the number of types of attributes of viewers of the television device is equal to or greater than a predetermined number;
A viewing information processing device characterized in that, when the attribute configuration estimated by the attribute configuration estimation unit is the third attribute configuration, the attribute configuration correction unit corrects the attribute configuration of the viewer of the television device to an attribute configuration in which one or more attributes included in the third attribute configuration, including at least the attribute with the lowest value of the attribute-specific existence probability identified by the attribute configuration estimation unit, are deleted . a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
The second step is configured to calculate a total value of the attribute-specific presence probability values, select a predetermined number of people based on the total value in order of the higher attribute-specific presence probability values, and estimate the number of people as an attribute of the constituent members of the viewers of the television device . a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device;
the predetermined abnormal attribute configuration includes a first attribute configuration in which an attribute configuration of a viewer of the television device consists solely of a child attribute, the child attribute being an attribute of a child under a first predetermined age;
The third step is a viewing information processing method characterized in that, when the attribute configuration estimated in the second step is the first attribute configuration, the attribute configuration of the viewer of the television device is corrected to an attribute configuration that includes the child attribute and attribute A, which is the attribute with the highest value of the attribute-specific existence probability identified in the second step among attributes other than the child attribute that satisfy the condition of being above a second predetermined age. a first step of acquiring viewing log information output from a television device including a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating when and which channel or broadcasting station television broadcast was viewed via the receiver;
a second step of estimating an attribute configuration of viewers of the television device from the viewing log information acquired in the first step,
the second step is a step of, when estimating the attribute composition of viewers of the television device, specifying, for each type of attribute, an attribute-specific presence probability, which is the probability that viewers having the attribute exist, and estimating, as an attribute of the constituent members of viewers of the television device, a predetermined attribute determined based on the specified attribute-specific presence probability;
a third step of determining, when the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, an attribute configuration of viewers of the television device by correcting the abnormal attribute configuration in accordance with an attribute-specific presence probability of viewers of the television device ;
the predetermined abnormal attribute configuration includes a second attribute configuration in which the number of types of attributes included in the attribute configuration of the viewer of the television device is zero;
The third step is a viewing information processing method characterized in that, when the attribute configuration estimated in the second step is the second attribute configuration, the attribute configuration of the viewer of the television device is corrected to an attribute configuration that includes attribute B, which is the attribute with the highest value of the attribute-specific existence probability identified in the second step, among the attributes that satisfy the condition of being a third predetermined age or older.