JP2023117392A - Viewing information processing device and viewing information processing method - Google Patents

Abstract

To provide a viewing information processing device capable of appropriately correcting an abnormal attribute configuration when an abnormal attribute configuration is estimated as an attribute configuration of a viewer of a television device from viewing log information of the television device.SOLUTION: A viewing information processing device 1 includes an attribute configuration estimation unit 12 that estimates an attribute configuration of a viewer of a television device 23 from viewing log information output from the television device 23, and an attribute configuration correction unit 13 that corrects the attribute configuration when the estimated attribute configuration is a predetermined abnormal attribute configuration. The attribute configuration correction unit 13 corrects the attribute configuration according to the attribute-specific existence probability specified by the attribute configuration estimation unit 12.SELECTED DRAWING: Figure 1

Description

The present invention relates to a viewing information processing apparatus and a viewing information processing method having a function of estimating a viewer's attribute structure from viewing log information obtained for each television device.

2. Description of the Related Art In recent years, television devices having a function of transmitting viewing data, including viewing log information indicating which channel's television broadcast was viewed and when, to a TV manufacturer or the like via an external network such as the Internet, have become widespread. In this specification, a device capable of receiving and viewing television broadcasts and having a function of transmitting viewing log information as described above is referred to as a television device.

Then, as seen in Patent Document 1, for example, from the viewing log information obtained from each television device, the viewers of the television device (more specifically, one or more persons who can view the television broadcast received via the television device) The application of the present application proposes a technique for estimating the attribute configuration of a viewer of the TV program using a model (mathematical model) that has undergone learning processing in advance. Here, the attribute composition of the viewer of the television device indicates which of the multiple types of attributes classified according to gender, age, etc., the viewer is made up of. is.

Patent No. 6433615

As seen in Patent Literature 1 above, when estimating the attributes of viewer members from viewing log information obtained from each TV device, there is a possibility that the attribute configuration of viewers obtained by the estimation is incorrect. It is possible to have an unusual attribute configuration with a high . For example, if the estimated audience attributes are only young children, or if the number of types of estimated audience attributes is zero (in other words, it is estimated that there are no TV device viewers). ), or if the number of types of estimated viewer attributes is too large, the estimated attribute configuration of the viewer is wrong, or it can be regarded as an abnormal attribute configuration that is highly likely to be. .

If the viewing log information of the TV device in which such an abnormal attribute composition is estimated and the attribute composition of the viewer are used to investigate the viewing situation such as the audience rating by attribute, the reliability of the investigation result will be damaged. There is a risk that

Here, it is conceivable to exclude the viewing log information of TV devices that are estimated to have an abnormal attribute composition from being used for viewing surveys. In that case, the total number of TV devices to be used should be sufficiently secured. There is a possibility that it will not be possible.

SUMMARY OF THE INVENTION The present invention has been made in view of such a background. It is an object of the present invention to provide a viewing information processing apparatus or a viewing information processing method capable of appropriately correcting an abnormal attribute configuration when a configuration is estimated.

In order to achieve the above object, the viewing information processing apparatus of the present invention is provided with a television broadcast receiver and a viewing log indicating which channel or broadcast station affiliated television broadcast was viewed through the receiver and when. a viewing log information acquisition unit that acquires viewing log information output from a television device that includes a viewing data output device capable of outputting information;
a viewing information processing apparatus comprising: an attribute configuration estimation unit for estimating an attribute configuration of a viewer of the television device from the viewing log information acquired by the viewing log information acquisition unit,
The attribute composition estimating unit is configured to, when estimating the attribute composition of the viewer of the television device, specify an attribute-specific existence probability, which is a probability that a viewer having the attribute exists, for each type of attribute. , configured to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of a constituent member of the viewer of the television device,
When the attribute configuration estimated by the attribute configuration estimating unit is a predetermined abnormal attribute configuration, the attribute configuration of the viewer of the television device is determined according to the presence probability by attribute of the viewer of the television device. , an attribute configuration correction unit that determines by correcting from the abnormal attribute configuration (first invention).

According to the first invention, when the attribute configuration estimated by the attribute configuration estimating unit regarding the viewer of the television device is an abnormal attribute configuration, according to the attribute-specific existence probability for each attribute included in the attribute configuration Since the attribute composition of the viewer of the television device is corrected, the attribute composition can be corrected in accordance with the attribute-specific existence probability based on the actual viewing log information related to the television device. Therefore, it is possible to appropriately correct the attribute configuration of the viewer of the television device from the estimated abnormal attribute configuration.

In the above-described first invention, the attribute structure estimation unit calculates the total value of the values of the existence probability by attribute, selects the predetermined number of people based on the total value in descending order of the value of the existence probability by attribute, and It is characterized in that it is configured to be estimated as an attribute of the constituent members of the viewers of the television device (second invention).

According to this, the attribute composition of the viewer of the television device is calculated based on the total value of the existence probability for each attribute included in the attribute composition estimated by the attribute composition estimation unit for the viewer of the television device. can be corrected easily and appropriately.
In the first invention, the attribute configuration estimation unit is configured to estimate an attribute whose attribute-specific existence probability value is equal to or greater than a predetermined threshold value as an attribute of a constituent member of a viewer of the television device. (third invention).

According to this, the attribute structure of the viewer of the television device is simplified based on the value of the existence probability for each attribute included in the attribute structure estimated by the attribute structure estimation unit for the viewer of the television device. and can be corrected appropriately.

In the present invention, with respect to arbitrary numerical values such as age and probability, "A or more" means being greater than A, or being greater than or equal to A. may Similarly, "A or less" may be either less than A, less than A, or equal to A.

In the first invention, the predetermined abnormal attribute configuration may include a first attribute configuration in which the viewer attribute configuration of the television device consists only of a child attribute that is an attribute of children under a first predetermined age. . In this case, when the attribute configuration estimated by the attribute configuration estimating unit is the first attribute configuration, the attribute configuration correcting unit adjusts the attribute configuration of the viewer of the television device to the child attribute and the child attribute. Among the attributes satisfying the condition of being at least a second predetermined age other than the above, the attribute configuration is corrected to include the attribute having the highest value of the existence probability for each attribute specified by the attribute configuration estimation unit. can be done (fourth invention).

According to this, the attribute configuration estimated as the first attribute configuration is not only the child attribute, but also the viewer of the television device among the attributes satisfying the second predetermined age or older attribute other than the child attribute. It can be corrected to include attribute A, which is the most likely attribute, ie, to have an attribute configuration similar to that of a typical household with children. Therefore, the attribute configuration of the viewer of the television device can be appropriately corrected from the first attribute configuration.

In the fourth aspect of the invention, when the attribute configuration estimated by the attribute configuration estimating unit is the first attribute configuration, the attribute configuration correction unit provides the child attribute and the second predetermined age or older condition. Among the attributes of age between the attributes that satisfy Alternatively, if there is an attribute with a value higher than the Pa, it is configured to correct the attribute configuration of the viewer of the television device so as to further include the existing attribute ( 5th invention).

According to this, even if the age attribute is between the child attribute and the attribute that satisfies the second predetermined age or older condition, if the attribute-specific existence probability value is high (of the television device). likely to be a viewer), the attribute can be further included in the viewer attribute configuration of the television device.

In the first aspect, the predetermined abnormal attribute configuration may include a second attribute configuration in which the number of attribute types included in the attribute configuration of the viewer of the television device is zero. In this case, when the attribute configuration estimated by the attribute configuration estimating unit is the second attribute configuration, the attribute configuration correcting unit adjusts the attribute configuration of the viewer of the television device to a third predetermined age or older. Among the attributes satisfying the condition, the attribute composition is corrected to include attribute B, which is the attribute with the highest value of the attribute-specific existence probability specified by the attribute composition estimation unit (sixth invention). .

When the sixth invention is combined with, for example, the fourth or fifth invention, the attribute that satisfies the third predetermined age or older condition is higher than the child attribute of the first predetermined age or younger. attribute of age. The third predetermined age is lower than the second predetermined age, is the same age as the second predetermined age, or is older than the second predetermined age. Either case is acceptable.

According to the sixth aspect, the attribute composition estimated as the second attribute composition is the attribute with the highest possibility of being the viewer of the television device among the attributes satisfying the third predetermined age or older condition. can be amended to include Therefore, the attribute configuration of the viewer of the television device can be appropriately corrected from the second attribute configuration.

In the sixth aspect, the attribute that satisfies the third predetermined age or older condition is an attribute other than a child attribute that is an attribute of a child under the first predetermined age, and the attribute configuration correcting unit is configured to: When the attribute configuration estimated by the attribute configuration estimating unit is the second attribute configuration, the value of the existence probability for each attribute specified by the attribute configuration estimating unit for the child attribute is The condition is that the value is the same as or higher than Pb, which is the value of the existence probability, and that the attribute B is at least a fourth predetermined age that is higher than the third predetermined age. It is possible to employ a mode in which the attribute configuration of the viewer of the television device is configured to further include the child attribute if the attribute is satisfied (seventh invention).

When the seventh invention is combined with, for example, the fourth invention or the fifth invention, the attribute satisfying the condition of being over the fourth predetermined age is higher than the child attribute under the first predetermined age. attribute of age. The fourth predetermined age may be lower than the second predetermined age, equal to the second predetermined age, or higher than the second predetermined age. .

According to the seventh aspect, when the attribute-specific existence probability value specified for the child attribute is high (when there is a high possibility that the child is a viewer of the television device), and the attribute B is the fourth A child attribute may be further included in the viewer attribute configuration of the television device if the attribute satisfies the condition of being of a predetermined age or older.

In the first to seventh inventions, the predetermined abnormal attribute configuration may include a third attribute configuration in which the number of attribute types 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 estimating unit is the third attribute configuration, the attribute configuration correction unit includes the attribute configuration of the viewer of the television device in the third attribute configuration. Among the attributes, one or more attributes including at least the attribute with the lowest value of the attribute-specific existence probability specified by the attribute configuration estimating unit may be corrected to an attribute configuration in which one or more attributes are deleted (eighth invention ).

According to the eighth aspect, the attribute configuration estimated as the third attribute configuration includes at least one attribute that is least likely to be the viewer of the television device among the attributes included in the third attribute configuration. Corrections can be made to remove the above attributes. Therefore, the attribute configuration of the viewer of the television device can be appropriately corrected from the third attribute configuration.

Also, in order to achieve the above object, the viewing information processing method of the present invention indicates a television broadcast receiver and when a television broadcast of which channel or broadcasting station affiliate was viewed via the receiver. a first step of acquiring viewing log information output from a television device including a viewing data output device capable of outputting viewing log information;
A viewing information processing method comprising: a second step of estimating an attribute configuration of a viewer of the television device from the viewing log information acquired in the first step,
In the second step, when estimating the attribute configuration of the viewers of the television device, for each attribute type, an attribute-specific existence probability, which is a probability that a viewer of the attribute exists, is identified, and the identified attribute is estimating a predetermined attribute determined based on the separate existence probability as an attribute of a constituent member of an audience of the television device;
When the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, the attribute configuration of the viewer of the television device is determined according to the presence probability by attribute of the viewer of the television device, It is characterized by further comprising a third step of correcting and determining from the abnormal attribute configuration (ninth invention).

According to the ninth aspect, when the attribute configuration estimated in the second step with respect to the viewer of the television device is an abnormal attribute configuration, the attribute configuration is determined according to the presence probability for each attribute, as in the first aspect. Therefore, the attribute configuration can be appropriately corrected from the abnormal attribute configuration.

The figure which shows the whole system in embodiment of this invention. FIG. 2 is a block diagram for explaining processing of an attribute configuration estimation unit of the viewing information processing apparatus shown in FIG. 1; 4 is a flowchart for explaining processing of an attribute configuration correction unit of the viewing information processing apparatus shown in FIG. 1; FIG. 2 is a block diagram for explaining processing of an attribute-based viewing information generating unit of the viewing information processing apparatus shown in FIG. 1;

One embodiment of the invention is described below with reference to FIGS. In the following description of the present embodiment, "greater than or equal to", "less than or equal to", and "less than" are used in their usual meanings. That is, A being greater than or equal to B means A≧B, A being less than or equal to B means A≦B, and A being less than B means A<B.

With reference to FIG. 1, the system described in the present embodiment has a viewing information processing device 1 that executes processing related to investigation of the viewing situation of television broadcasting. The viewing information processing apparatus 1 is composed of, for example, one or more computers. The computer includes a processor such as a CPU, a memory (storage device), an interface circuit, a communication device, and the like (not shown). Then, the viewing information processing apparatus 1 connects a plurality of device-specific households 20 belonging to an area to be surveyed for viewing conditions (for example, an area in units of prefectures, or an area including a plurality of prefectures such as the Kanto area and the Kinki area). , it is possible to acquire viewing data regarding TV broadcast viewing in each device-specific household 20 . The viewing data is data including viewing log information or the like indicating when a television broadcast of which channel (or which broadcasting station series) was viewed.

Here, in this embodiment, each “device-specific household 20” includes a television device 23 including a television broadcast receiver 21 and a viewing data output device 22 capable of outputting viewing data relating to viewing of the television broadcast. , means a household comprising one or more viewers of the television device 23 (more specifically, one or more viewers who can view the television broadcast received by the receiver 21 of the television device 23). In other words, the device-specific household 20 is configured as a set of the television device 23 and the viewer of the television device 23 .

In this case, one device-specific household 20 is associated with each television device 23 . Therefore, even if a certain dwelling unit is provided with a plurality of television devices 23 , one device-specific household 20 is associated with each of the plurality of television devices 23 . In this case, each device-specific household 20 corresponding to each of the plurality of television devices 23 may include the same members as viewers. In other words, one or more members of any device-specific household 20 may be members of other device-specific households 20 .

The television device 23 of each device-specific household 20 may consist of a television or may consist of a television and a recording device connected thereto. The viewing data output device 22 of the television device 23 is composed of a processor such as a microcomputer (not shown), a memory, an interface circuit, a communication device, and the like. The viewing data output device 22 is capable of detecting the channel of the television broadcast viewed via the television device 23 including it, and the date and time when the television broadcast of the channel was viewed (more specifically, the channel It is possible to detect the viewing start and viewing end date and time of the television broadcast), and from this detection information, the viewing log information of the device-specific household 20 (hereinafter referred to as device-specific viewing log information) can be generated. is possible. The viewing log information by device is, in other words, the viewing log information of all members included in the household 20 by device.

Then, the viewing data output device 22 transmits the viewing data including the generated device-specific viewing log information to the television or recording device constituting the television device 23 via an external network NW configured by the Internet, a telephone line network, or the like. It is possible to periodically (or in response to a request from the server) transmit to the manufacturer's server (not shown). In this case, the viewing data transmitted from the viewing data output device 22 includes device-specific viewing log information as well as location area information (for example, postal code number) registered in advance in the television or recording device that constitutes the television device 23. 3-digit information) and identification information of the television device 23 are included. The identification information of the television device 23 can also be used as identification information of the device-specific household 20 having the television device 23 .

The viewing information processing apparatus 1 includes a viewing data acquiring unit 11, an attribute configuration estimating unit 12, an attribute configuration correcting unit 13, and an attribute-based viewing information generating unit as functions realized by the installed hardware configuration and programs (software configuration). A viewing condition estimation unit 15 is provided. Hereinafter, the overall processing of the viewing information processing apparatus 1 will be described together with the details of each of these functional units.

The viewing data acquisition unit 11 can communicate with a server (not shown) of the manufacturer of each television device 23 of each device-specific household 20. Viewing data can be obtained. The viewing data acquisition section 11 functions as a viewing log information acquisition section in the present invention.

When the viewing information processing apparatus 1 can communicate with the viewing data output apparatus 22 of the device-specific household 20, the viewing data acquisition unit 11 acquires the viewing data of the television device 23 of the device-specific household 20 from the television. It may be obtained directly from the viewing data output device 22 of the device 23 . Also, the viewing information processing apparatus 1 may acquire viewing data of the television device 23 from each manufacturer via an appropriate storage device.

The viewing information processing apparatus 1 uses the viewing data acquired by the viewing data acquisition unit 11 for each device-specific household 20 in the survey target area to generate an attribute configuration estimation unit 12, an attribute configuration correction unit 13, and an attribute-specific viewing information generation unit. The processes of the unit 14 and the viewing situation estimation unit 15 are sequentially executed.

The attribute structure estimating unit 12, from the device-specific viewing log information for a predetermined period (for example, one month, several months, one year, etc.) output from the television device 23 of each device-specific household 20, the device-specific household 20 (in other words, the attribute configuration of the viewers of the television devices 23 in the device-specific household 20).

Here, the attributes of the constituent members (viewers of the television devices 23) of each device-specific household 20 are classified into a plurality of types of attributes according to the age and sex of the constituent members, for example. For example, children under y1, men over y1 and under y2, women over y1 and under y2, men over y2, women over y2, etc. being classified. Hereinafter, the number of attribute types is assumed to be N, and each of the N types of attributes is appropriately described as At(i) (i=1, 2, . . . , N). Note that the attributes of members of each device-specific household 20 can be classified according to various parameters, such as occupation and educational background, in addition to age and gender.

As shown in FIG. 2, the attribute composition estimating unit 12 estimates the attribute composition of the constituent members of each device-specific household 20 using a first model subjected to machine learning processing in advance. Here, the first model is designed to be able to presume the attribute composition of the constituent members of any device-specific household 20 from the device-specific viewing log for a predetermined period of the device-specific household 20 belonging to the research target area. This is a model that has undergone learning processing. In the machine learning process for the first model, viewing log information for each sample household obtained from a people meter or the like installed in each of a plurality of sample households whose attribute composition is known and the attribute composition for each sample household are learned. Used as data. A known algorithm can be used as an algorithm for the machine learning processing.

In this case, in the present embodiment, the first model is based on N types of attributes At(i) (i=1, 2, . Each of them is configured to be able to specify an attribute-specific existence probability, which is the probability that a member of each attribute At(i) exists in the device-specific household 20 . As such a first model, for example, a model similar to the mathematical model for determining household composition in Patent Document 1 can be adopted. However, the first model may be any other type of model as long as it can specify attribute-specific existence probabilities (or similar index values) from device-specific viewing log information for a predetermined period of device-specific households 20. There may be.

Then, the attribute composition estimating unit 12 selects an attribute whose presence probability by attribute specified by the first model is equal to or greater than a predetermined threshold value (for example, 0.5 or more) as an attribute of the members of the household 20 by device. presume. As an example, FIG. 2 illustrates the processing of the attribute configuration estimation unit 12 for one device-specific household 20 (the device-specific household 20 whose identification information ID is x1 in FIG. 2). In the block diagrams shown in FIG. 2 and FIG. 4, which will be described later, a processing execution unit is indicated by a thick line frame, and a data output unit or data is indicated by a thin line frame.

In the example shown in FIG. 2, based on the device-specific viewing log information for a predetermined period of the device-specific household 20 whose identification information ID is x1, the presence probability for each attribute of members of the device-specific household 20 is shown by the first model. is specified as In this example, out of N types of attributes At(i) (i=1, 2, . is equal to or higher than the threshold value (here, for example, 0.5), and the attribute-specific existence probabilities of the other attributes At(2), At(4) to At(N-1) are smaller than the threshold value. ing.

In this case, the attribute composition estimating unit 12 determines that the attributes At(1), At(3), and At(N) whose attribute-specific existence probabilities are equal to or greater than the threshold are members of the device-specific household 20 (viewing of the television device 23). Attributes At(2), At(4) to At(N-1) whose attribute-specific existence probabilities are smaller than the threshold are assumed to be attributes of members of the device-specific household 20. presume. As a result, the attribute composition of the constituent members of the device-specific household 20 is estimated.

In this way, the attribute composition estimating unit 12 determines whether or not the attribute-specific existence probability specified by the first model for each device-specific household 20 is equal to or greater than a predetermined threshold value, and the constituent members (viewers) of each attribute are By estimating whether or not they exist in the device-specific household 20, the attribute composition of the constituent members of the device-specific household 20 is estimated.

After executing the process of the attribute structure estimation unit 12 , the viewing information processing apparatus 1 executes the process of the attribute structure correction unit 13 next. The attribute configuration correcting unit 13 determines that the attribute configuration estimated by the attribute configuration estimating unit 12 for each device-specific household 20 is a predetermined abnormal attribute configuration that is highly likely to be wrong. Second, it is a functional unit that corrects the estimation result of the attribute composition of the constituent members of the device-specific household 20 .

Here, in this embodiment, for example, three types of attribute configurations (first to third attribute configurations) are predetermined as predetermined abnormal attribute configurations. In the first attribute configuration, the constituent members estimated by the attribute configuration estimation unit 12 for the device-specific household 20 are under y1 years old ( Or y1 years old or younger), it is an attribute configuration consisting only of constituent members with the attribute of children (hereinafter, this attribute will be referred to as the attribute of At(1)). In other words, the first attribute configuration is, among the attribute-specific existence probabilities specified for each of N types of attributes At(i) (i=1, 2, . . . , N), less than y1 (or y1 This is an attribute configuration when only the attribute-specific existence probability corresponding to the attribute At(1) (hereinafter simply referred to as the child attribute At(1)) of a child of age or younger is equal to or greater than a predetermined threshold. The age y1 corresponds to the first predetermined age in the present invention.

In addition, the second attribute configuration indicates which of the N types of attributes At(i) (i=1, 2, . (ie, the total number of attributes included in the attribute configuration is zero). In other words, the second attribute configuration is such that all attribute-specific existence probabilities specified for each of N types of attributes At(i) (i=1, 2, . . . , N) are higher than a predetermined threshold value. This is the attribute configuration when it becomes a small 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 (eg, 5 or 6) or more types of attributes. In other words, the third attribute configuration is, among the attribute-specific existence probabilities specified for each of the N types of attributes At(i) (i=1, 2, . . . , N), a predetermined number or more of types This is an attribute configuration when the attribute-specific existence probability corresponding to each attribute is equal to or greater than a predetermined threshold value. In the following description, the predetermined number related to the third attribute configuration is expressed as a predetermined number nx.

These first to third attribute configurations are attribute configurations that can be regarded as not existing in reality (eg, first attribute configuration or second attribute configuration), or other than the first to third attribute configurations It is an attribute configuration (for example, a third attribute configuration) that can be considered to exist rarely compared to attribute configurations.

The attribute composition correcting unit 13 receives the attribute composition estimated by the attribute composition estimating unit 12 for each device-specific household 20 and the existence probability by attribute specified by the first model of the attribute composition estimating unit 12. . Then, the attribute configuration correction unit 13 determines whether or not the estimated attribute configuration is an abnormal attribute configuration of any of the above first to third attribute configurations for each device-specific household 20, and determines whether the estimated attribute configuration is abnormal. If the attribute configuration obtained is any abnormal attribute configuration, correct the attribute configuration.

Specifically, the attribute configuration correction unit 13 performs the processing shown in the flowchart of FIG. Correct the attribute configuration for the device-specific household 20 that has changed.

In STEP 1, the attribute configuration correction unit 13 determines whether or not the attribute configuration estimated for each device-specific household 20 is the third attribute configuration (attribute configuration including a predetermined number nx or more types of attributes). . When the determination result is affirmative (when the estimated attribute configuration is the third attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP2.

In this STEP 2, the attribute configuration correction unit 13 calculates the attribute-specific existence probabilities specified for the device-based household 20 from the attribute configuration of the device-based household 20 (the device-based household 20 for which the determination result in STEP 1 is affirmative). The total number of attributes included in the attribute configuration of the device-specific household 20 is reduced to a total number less than a predetermined number nx by correcting the attribute configuration to remove the attributes selected in response to .

Specifically, the attribute configuration correcting unit 13 determines, out of the attribute-based existence probabilities corresponding to each of a plurality of attributes (at least a predetermined number nx types of attributes) included in the attribute configuration estimated for the device-based household 20, , Select one or more types of attributes with lower probability values (probability values closer to the lowest probability value than the highest probability value (including the lowest probability value)) as attributes to be deleted, and select By deleting the attribute to be deleted from the attribute configuration of the device-specific household, the attribute configuration is corrected so that the total number of attributes after the deletion is less than the predetermined number nx. In this case, among the plurality of attributes included in the estimated attribute configuration, an attribute with a lower attribute-specific existence probability value (probability value) is 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 selects attributes (attributes with relatively low values of existence probability by attribute (probability values) among a plurality of attributes (at least a predetermined number of nx types of attributes) included in the estimated attribute configuration ( Attributes with lower probability values) are selected in order from the one with the lowest probability value and deleted until the total number of attributes after deletion is less than a predetermined number nx. In this case, in each iteration, if there are multiple attributes with the lowest value of existence probability for each attribute among the attributes included in the attribute configuration before deletion, those multiple attributes are deleted.

As an example, the predetermined number nx is, for example, "5", and the attribute configuration estimated by the attribute configuration estimation unit 12 for a certain device-specific household 20 is, for example, from seven types of attributes At(1) to At(7). It is an attribute configuration consisting of At(1) to At(7), and the value of each attribute's existence probability is 0.63 (:At(1)), 0.53 (:At(2)), 0.53 (:At(3 )) 0.70 (:At(4)), 0.80 (:At(5)), 0.57 (:At(6)), 0.57 (:At(7)).

In this case, the two types of attributes At(2) and At(3) with the lowest value of the existence probability by attribute are deleted from the attributes At(1) to At(7). Two types of attributes At(6) and At(7) with low values of are deleted. As a result, the attribute configuration in the device-specific household 20 is corrected to an attribute configuration including three types of attributes At(1), At(4), and At(5).

In this embodiment, the process of STEP 2 (the process of correcting 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 so that the number of attribute types included therein is less than the predetermined number nx.

The attribute configuration correction unit 13 executes the process of STEP3 when the determination result of STEP1 is negative (when the estimated attribute configuration is not the third attribute configuration). Also, the attribute configuration correction unit 13 executes the process of STEP3 after executing the process of STEP2. In STEP 3, the attribute configuration correction unit 13 determines whether or not the estimated attribute configuration is the first attribute configuration (attribute configuration including only the child attribute At(1)).

Here, in the present embodiment, the attribute configuration after correction by the processing of STEP 2 may become the first attribute configuration. For example, the attribute configuration (=third attribute configuration) before correction by the processing of STEP 2 includes the child attribute At(1), and the attribute-specific existence probability of other attributes in the attribute configuration before correction is lower than the value of the attribute-specific existence probability of the child attribute At(1) and both are the same value, the attribute configuration after correction by the processing in STEP 2 is the child attribute At(1 ), that is, the first attribute configuration. Therefore, in the present embodiment, the attribute configuration correction unit 13 executes the determination process of STEP3 not only when the determination result of STEP1 is negative, but also after the process of STEP2 is performed.

If the determination result of STEP3 is affirmative (if the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration after correction in STEP2 is the first attribute configuration), the attribute configuration correction unit 13 executes the processing of STEP4.

In STEP 4, the attribute configuration correction unit 13 adds the attribute configuration of the device-specific household 20 (the device-specific household 20 for which the determination result in STEP 3 is affirmative) to the child attribute At(1), The attribute configuration is corrected to include one or more attributes selected according to the attribute-specific existence probability specified for the household 20 .

Specifically, the attribute composition correction unit 13 selects N types of attributes At(i) (i=1, 2, . . . , N ), select the attribute with the highest probability of existence by attribute from among the attributes that satisfy the condition of being at least a predetermined age (adult age) as the attribute to be added, and select the attribute to be added for each device The attribute configuration of the device-specific household 20 is corrected so as to be added to the attribute configuration of the household 20 (attribute configuration consisting only of the child attribute At(1)). In this case, if there are a plurality of attributes with the highest attribute-specific existence probability values, these attributes are added to the attribute configuration of the device-specific household 20 . In addition, the predetermined age corresponds to the second predetermined age or the fourth predetermined age in the present invention.

For example, the attribute configuration estimated by the attribute configuration estimating unit 12 for a device-specific household 20 is the first attribute configuration, and the attributes that satisfy the condition of a predetermined age or older (adult age) are At(4) to At(7). ), and the value of the existence probability for each attribute of At(4) to At(7) is 0.45 (:At(4)), 0.36(:At(5)), 0.45(: At(6)), 0.28 (:At(7)). In this case, among At(4) to At(7), the attributes with the highest value of existence probability by attribute are At(4) and At(6). ) 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 the device-specific household 20 whose attribute configuration after correction in STEP 2 is the first attribute configuration. Supplementally, in the processing of STEP 4, the attribute configuration after correction may become an attribute configuration (=third attribute configuration) including the types of attributes equal to or greater than the predetermined number nx. In addition, in the processing of STEP 4, in addition to attributes satisfying the condition of being at least a predetermined age (adult age) (hereafter referred to as adult attributes), attributes of age between the adult attribute and the child attribute At(1) are included. The attribute configuration of the device-specific household 20 may be corrected so that the attribute configuration of the device-specific household 20 may be further included (hereinafter referred to as the youth attribute).

Specifically, the value of the attribute-specific existence probability for a certain youth attribute is a value equal to or greater than the highest attribute-specific existence probability value Pa among the respective attribute-specific existence probabilities of the adult attribute (is it the same value as Pa, or a value higher than Pa), not only the attribute whose attribute-specific existence probability value is Pa, but also the young attribute having an attribute-specific existence probability value of Pa or more is added to the attribute configuration of the device-specific household 20 may be added.

For example, the adult attribute is four types of attributes At(4) to At(7), and the youth attribute is two types of attributes At(2) and At(3). The values of the attribute-specific existence probabilities of At(2) to At(7) in the device-specific household 20) estimated as the attribute composition are 0.49 (:At(2)), 0.30 (:At(3), 0.45 (:At(4)), 0.36 (:At(5)), 0.45 (:At(6)), 0.28 (:At(7)) In this case, the adult attribute At(4) Attributes At(4) and At(6) with the highest attribute-specific existence probability values among ~At(7), and attributes At(4) and At(4) and At(3) among youth attributes At(2) and At(3) Attribute At(2) having a higher attribute-specific existence probability value than (6) is selected as an attribute to be added, and is added to the attribute configuration of the device-specific household 20. As a result, the attribute of the device-specific household 20 is selected. The configuration 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 (the process 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 first attribute configuration is the attribute configuration estimated by the attribute configuration estimating unit 12 or the attribute configuration after correction in STEP 2 is the attribute other than At(1) ( The attribute configuration is corrected to include an adult attribute). Note that when the process of STEP 4 is executed, the process of the attribute composition correction unit 13 for the device-specific household 20 whose attribute composition has been corrected by the process ends (and thus the attribute composition of the device-specific household 20 is determined). ).

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 , the processing of STEP5 is executed. In STEP 5, the attribute configuration correction unit 13 determines whether or not the estimated attribute configuration is the second attribute configuration (attribute configuration in which the total number of included attributes is zero).

Here, in the present embodiment, the attribute configuration after correction by the processing of STEP 2 may become the second attribute configuration. For example, all of the attribute-specific existence probability values of a plurality of attributes (the number of types of attributes 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 the same value. In this case, the attribute configuration after correction by the processing of STEP 2 becomes the attribute configuration in which the total number of attributes included therein is zero, ie, the second attribute configuration. For this reason, in the present embodiment, the attribute configuration correction unit 13 not only determines that the determination results in STEP 1 and 3 are negative, but also determines that the determination result in STEP 3 becomes negative after executing the process in STEP 2. Also in this case, the judgment processing of STEP5 is executed.

If the determination result in STEP5 is affirmative (if the attribute configuration estimated by the attribute configuration estimation unit 12 or the attribute configuration corrected in STEP2 is the second attribute configuration), the attribute configuration correction unit 13 Execute the processing of STEP6.

In STEP 6, the attribute composition correction unit 13 converts the attribute composition of the device-specific household 20 (the device-specific household 20 for which the determination result in STEP 5 is affirmative) into the attribute-specific existence probabilities specified for the device-specific household 20. Correct the attribute configuration to include one or more attributes selected according to.

Specifically, the attribute composition correction unit 13 selects N types of attributes At(i) (i=1, 2, . . . , N ), from among the attributes At(2) to At(N) excluding the child attribute At(1), the attribute with the highest value of existence probability by attribute is selected as the attribute to be added, and the selected attribute to be added The attribute configuration of the device-specific household 20 is corrected so that the attribute configuration of the device-specific household 20 is the attribute configuration of the device-specific household 20 . Note that the attributes At(2) to At(N) excluding the child attribute At(1) are, in other words, ages equal to or higher than the predetermined age y1 (or older than y1) related to the child attribute At(1). and the predetermined age y1 corresponds to the third predetermined age in the present invention.

In this case, among the attributes At(2) to At(N) excluding the child attribute At(1), if there are multiple attributes with the highest value of existence probability for each attribute, then the attribute is composed of those multiple attributes. The attribute configuration of the device-specific household 20 is corrected so as to obtain the attribute configuration.

For example, the attribute configuration estimated by the attribute configuration estimating unit 12 for a certain device-specific household 20 is the second attribute configuration. are both 0.47, and the values of the attribute-specific existence probabilities of the other attributes At(2), At(5) to At(N) are all lower than 0.47. In this case, among the attributes At(2) to At(N), the attributes with the highest value of existence probability by attribute are the two attributes At(3) and At(4). 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 composed of these attributes to be added At(3) and At(4).

The same applies to the device-specific household 20 whose attribute configuration after correction in STEP 2 is the second attribute configuration. Supplementally, in the processing of STEP 6, the attribute configuration after correction may become an attribute configuration (=third attribute configuration) including the types of attributes equal to or greater than the predetermined number nx. Note that in the processing of STEP 6, the attribute configuration of the device-specific household 20 may be corrected so that not only attributes other than the child attribute At(1) but also the child attribute At(1) may be included.

Specifically, the attribute-specific existence probability value for the child attribute At(1) is equal to or higher than the highest attribute-specific existence probability value Pb among the respective attribute-specific existence probabilities other than the child attribute At(1) ( Pb or a value higher than Pb) and the attribute whose existence probability value by attribute is Pb includes an adult attribute, only the attribute whose existence probability value by attribute is Pa Instead, the child attribute At(1) may be added to the attribute configuration of the device-specific household 20 .

For example, the adult attribute is four types of attributes At(4) to At(7), and the youth attribute is two types of attributes At(2) and At(3). The attribute-specific existence probability of the child attribute At(1) in the device-specific household 20) estimated to be the attribute composition is 0.48, and among the attributes At(2) to At(7), the youth attribute At(3) and the adult The value of existence probability by attribute for attribute At(4) is 0.47, and the value of existence probability by attribute for other attributes At(2), At(5) to At(7) are all Assume lower than 0.47. Among the attributes At(2) to At(7) other than the child attribute At(1), the attributes At(3) and At(4) with the highest value of existence probability by attribute include the adult attribute At(4). and the value of the attribute-specific existence probability of the child attribute At(1) is higher than the values of the attribute-specific existence probabilities 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. ), the attribute configuration of the device-specific household 20 is corrected.

Note that if the attribute with the highest attribute-specific existence probability value among the attributes At(2) to At(7) is only the youth attribute (for example, At(3)), the child attribute At(1 ) is greater than or equal to the value of the attribute-specific existence probability of At(3), the child attribute At(1) is not selected as an attribute to be added. Therefore, the corrected attribute configuration is prevented from becoming an attribute configuration consisting only of the child attribute At(1) and the youth attribute.

In this embodiment, the process of STEP 6 (the process of correcting 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 estimating unit 12 or the attribute configuration corrected in STEP 2 is corrected to include one or more attributes. be done.

In this embodiment, the processing of the attribute configuration correction unit 13 is executed as described above. As a result, in the device-specific household 20 whose attribute configuration estimated by the attribute configuration estimating unit 12 is the first attribute configuration, in addition to the child attribute At(1), the condition of being an adult over a predetermined age is set. The attribute configuration of the device-specific household 20 can be corrected so as to include at least the attribute (adult attribute) that satisfies the requirement. In this case, the attribute added other than the child attribute At(1) has a higher attribute-specific existence probability value than the other attributes. It is possible to prevent deviation from the actual attribute configuration as much as possible.

In addition, in the device-specific household 20 whose attribute configuration estimated by the attribute configuration estimation unit 12 is the second attribute configuration, the device-specific household 20 includes one or more attributes other than the child attribute At(1). Twenty attribute configurations can be corrected. In this case, since the attributes to be included in the attribute configuration after correction are attributes having a higher attribute-specific existence probability value than the other attributes, the attribute configuration after correction is the actual attribute of the device-specific household 20. Deviation from the configuration can be prevented as much as possible. Furthermore, since the attributes to be included in the attribute structure after correction include at least attributes other than the child attribute At(1), the attribute structure after correction consists only of the attributes of the child At(1) (=first attribute configuration).

In addition, in the device-specific household 20 in which the attribute configuration estimated by the attribute configuration estimation unit 12 is the third attribute configuration, basically, the attribute configuration is such that the number of types of attributes is less than the predetermined number nx. In addition, the attribute configuration of the device-specific household 20 can be corrected. In this case, since the attribute to be deleted from the attribute configuration before correction has a lower attribute-specific existence probability value than the other attributes, the attribute configuration after correction corresponds to the actual attribute of the device-specific household 20. Deviation from the configuration can be prevented as much as possible.

Although the attribute configuration after correction by the processing in STEP 4 or 6 may become a third attribute configuration including attributes of a predetermined number nx or more, the frequency of occurrence in such cases is generally low. Also, the third attribute configuration is an attribute configuration that is more likely to actually exist than the first attribute configuration and the second attribute configuration. Therefore, in this embodiment, when the attribute configuration after correction by the processing of STEP 4 or 6 becomes the third attribute configuration, that attribute configuration is maintained.

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

The attribute-based viewing information generating unit 14 is as shown in FIG. From the viewing log information of each device-specific household 20 for a predetermined period, attribute-specific viewing information for each component attribute of each device-specific household 20 is estimated using a second model subjected to machine learning processing in advance.

Here, in the present embodiment, the second model is a model created for each of N types of attributes At(i) (i=1, 2, . . . , N). In the second model corresponding to each attribute At(i), the channel (or broadcasting station series) of the broadcasting to be investigated and the date and time data indicating the date and time (month, day of the week, time zone, etc.) of the broadcast are specified. Then, from the device-specific viewing log for a predetermined period of the device-specific household 20 to which the attribute At(i) belongs, the probability (hereinafter , attribute-specific viewing probability), the model has undergone machine learning processing in advance.

Note that in FIG. 2, from the viewing log information of the device-specific household 20 having a member with a certain attribute At(n), the second model corresponding to the attribute At(n) is used to generate channel (or broadcasting station series) and date/time data. It is shown that the attribute-specific viewing probability is specified, which is the probability that the constituent member of the attribute At(n) has viewed the investigation target broadcast defined by . The same is true for second models corresponding to other attributes.

In the machine learning process for the second model as described above, viewing log information for each member of each sample household obtained from people meters installed in each of a plurality of sample households whose attribute configuration is known and each sample household Each attribute configuration is used as learning data. A known algorithm can be used as an algorithm for the machine learning processing. Also, as such a second model, for example, a model similar to the mathematical model for individual viewing determination in Patent Literature 1 can be adopted. However, in the second model, for each attribute of members of the device-specific household 20, the attribute-specific viewing probability (or similar Any other form of model may be used as long as it can specify the index value).

Then, the attribute-specific viewing information generation unit 14 generates the attribute-specific viewing probability specified by the second model for each attribute of each constituent member of the device-specific household 20 and the device-specific viewing log information of the device-specific household 20 ( It is determined (estimated) whether or not the constituent member of each attribute has viewed the survey target broadcast from the device-specific viewing log information including the log information of the broadcast date and time of the survey target broadcast.

Specifically, the attribute-specific viewing information generation unit 14 sets the value (probability value) of the attribute-specific viewing probability specified by the second model for the survey target broadcast to a predetermined value for each attribute of the members of the device-specific household 20. It is above the threshold (for example, 0.5 or more), and the broadcast date and time of the survey target broadcast (currently, there was actual viewing on the TV broadcast channel (or broadcasting station affiliate)) If the condition that it is confirmed from the viewing location information of the household 20 is satisfied, it is determined that the survey target broadcast was viewed by the constituent member of the attribute, and if the condition is not satisfied, the constituent member of the attribute. It is determined that there was no viewing of the survey target broadcast by

For example, in FIG. 4, the viewing probability by attribute related to the attribute Ar(n) is equal to or greater than a predetermined threshold, and at the date and time indicated by the specified date and time data of the television broadcast, the channel (or broadcast station) of the television broadcast If it is confirmed from the device-specific viewing location information that there was actual viewing in the affiliate), the attribute At(n) shows the situation in which it is determined that there was viewing of the investigation target broadcast by the constituent personnel. there is

In the present embodiment, the process of the attribute-based viewing information generation unit 14 is executed as described above. As a result, for each device-specific household 20, for each attribute included in the attribute configuration determined by the attribute configuration correction unit 13, whether the survey target broadcast for which the channel (or broadcasting station series) and date and time data are specified has been viewed Attribute-specific viewing information is generated that indicates whether or not.

The viewing information processing apparatus 1 then executes the processing of the viewing situation estimation unit 15 . The viewing situation estimating unit 15, for each device-based household 20 in the survey target area, based on the attribute-based viewing information obtained by the processing of the attribute-based viewing information generating unit 14, determines the number of device-based households 20 in the survey target area. This is a processing unit that estimates the viewing status data (data corresponding to the audience rating) of the survey target broadcast for each attribute of the entire constituent members.

In this case, in this embodiment, M(i) is the total number of constituent members for each attribute At(i) (i=1, 2, . In addition, if the total number of members determined to have viewed the survey target broadcast for each attribute At(i) is denoted by m(i), the viewing situation estimation unit 15 calculates for each attribute At(i) is calculated by the following equation (1), for example.

Viewing situation data for attribute At(i)=m(i)/M(i) (1)

This makes it possible to obtain viewing status data as data corresponding to the audience rating for each attribute in the research target area.

According to the viewing information processing apparatus 1 of the present embodiment described above, the attribute configuration of each device-based household 20 estimated by the attribute configuration estimation unit 12 from the device-based viewing log information of each device-based household 20 in the survey target area is If the attribute configuration is any one of the first attribute configuration, the second attribute configuration, and the third abnormal attribute configuration, the attribute configuration is corrected according to the attribute-specific existence probability. By this correction, the survey target area does not include the device-specific households 20 with the first attribute configuration or the second attribute configuration, and the device-specific households 20 with the third attribute configuration fall within a sufficiently small number of households. can.

Therefore, it is possible to identify the attribute composition of each device-specific household in the research target area so as to ensure high reliability. As a result, the reliability of the attribute-based viewing information of each device-based household by the processing of the attribute-based viewing information generating unit 14 and the reliability of the viewing situation data by the processing of the viewing situation estimating unit 15 can be improved. As a result, it is possible to properly survey the viewing situation in the survey target area.

It should be noted that the present invention is not limited to the embodiments described above, and other embodiments can be employed. Some other embodiments are illustrated below.
In the above-described embodiment, the attribute composition estimating unit 12 selects an attribute whose presence probability by attribute specified by the first model is equal to or greater than a predetermined threshold value (for example, 0.5 or more) as a member of the household 20 by device. Although the case of estimating as an attribute has been described, the present invention is not limited to this. Specifically, if the attribute composition estimating unit 12 is a predetermined attribute determined based on the existence probability by attribute, the total value of the values of the existence probability by attribute is calculated, and the predetermined number of people based on the total value is calculated. Select from the highest value of the existence probability by attribute (for example, if the total value is 1.8 exceeding 1, one person with the highest value, 2.5 exceeding 2, the highest value 2 from the first), the selected attribute may be estimated as the attribute of the constituent members of the device-specific household 20 .
In this case, if there are multiple people with the same attribute-specific existence probability value, it is conceivable to randomly select which subject to prioritize. Priorities may be set according to information such as the TV manufacturer and the number of minutes watched.
In addition, a method may be adopted in which the estimated number of people is defined by attribute (not limited to the total value of the existence probability values by attribute), and the attribute of the constituent factor is determined in descending order of the existence probability value.
Further, in the above-described embodiment, three types of attribute configurations, namely, the first attribute configuration, the second attribute configuration, and the third attribute configuration, are used as the abnormal attribute information. configuration, and the third attribute configuration may be excluded from the abnormal attribute configuration. Alternatively, for example, whether or not to exclude the third attribute configuration is selected for each survey target area, or the third attribute configuration is selected among the entire device-specific households 20 whose attribute configurations are estimated by the attribute configuration estimation unit 12. The selection may be made according to whether or not the percentage of households 20 classified by device is higher than a predetermined value.

Furthermore, in the above embodiment, when the estimated attribute configuration is the third attribute configuration, the attribute configuration has the same attribute-specific existence probability value, and the value is higher than the other attributes. When multiple low-level attributes were included, all of the multiple attributes were selected as attributes to be deleted. However, in this case, for example, within the range where the total number of attributes included in the attribute configuration after correction (after deletion of one or more attributes) is less than the predetermined number nx, the presence probability by attribute is satisfied. Attributes to be deleted may be selected at random or in a predetermined order of priority from the plurality of attributes having the same value.

Further, in the above-described embodiment, when the estimated attribute configuration or the attribute configuration after correction from the third attribute configuration is the first attribute configuration, the attributes satisfying the condition that the attribute has the age of an adult. Among them, when there are multiple attributes that have the same value of the existence probability for each attribute and the value is higher than other attributes, the attribute is added to the attribute configuration as a whole. Corrected the configuration. However, in this case, for example, within the range where the total number of attributes included in the attribute configuration after correction (after adding one or more attributes) is less than the predetermined number nx, the presence probability by attribute is satisfied. Attributes to be added may be selected at random or in a predetermined order of priority from the plurality of attributes having the same value. This is the same when adding a new attribute 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 apparatus 1 includes the attribute-based viewing information generating unit 14 and the viewing situation estimating unit 15, but the viewing information processing apparatus of the present invention includes the attribute-based viewing information generating unit 14 and The viewing state estimation unit 15 may not be provided. In this case, the attribute-based viewing information generation unit 14 and the viewing situation estimation unit 15 may be provided in a device separate from the viewing information processing device. Further, the attribute configuration estimated by the attribute configuration estimation unit 12 or corrected by the attribute configuration correction unit 13 may be used in separate processing from the attribute-based viewing information generation unit 14 and the viewing situation estimation unit 15. good.

Reference Signs List 1 viewing information processing device 11 viewing data acquisition unit (viewing log information acquisition unit) 12 attribute configuration estimation unit 13 attribute configuration correction unit 21 receiver 22 viewing data output device 23 television device.

Claims (9)

Output from a television device comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channel or broadcasting station affiliated television broadcast was viewed through the receiver and when a viewing log information acquisition unit that acquires viewing log information;
a viewing information processing apparatus comprising: an attribute configuration estimation unit for estimating an attribute configuration of a viewer of the television device from the viewing log information acquired by the viewing log information acquisition unit,
The attribute composition estimating unit is configured to, when estimating the attribute composition of the viewer of the television device, specify an attribute-specific existence probability, which is a probability that a viewer having the attribute exists, for each type of attribute. , configured to estimate a predetermined attribute determined based on the attribute-specific existence probability as an attribute of a constituent member of the viewer of the television device,
When the attribute configuration estimated by the attribute configuration estimating unit is a predetermined abnormal attribute configuration, the attribute configuration of the viewer of the television device is determined according to the presence probability by attribute of the viewer of the television device. A viewing information processing apparatus, comprising: an attribute configuration correction unit that determines by correcting the abnormal attribute configuration. The viewing information processing device according to claim 1,
The attribute configuration estimating unit calculates the total value of the values of the existence probability by attribute, selects a predetermined number of people based on the total value in descending order of the value of the existence probability by attribute, and selects the number of viewers of the television device A viewing information processing apparatus characterized by being configured to be estimated as an attribute of a constituent member. The viewing information processing device according to claim 1,
The attribute configuration estimating unit is configured to estimate an attribute whose value of the existence probability for each attribute is equal to or greater than a predetermined threshold as an attribute of a constituent member of the viewer of the television device. Information processing equipment. The viewing information processing device according to claim 1,
the predetermined anomalous attribute configuration includes a first attribute configuration in which the attribute configuration of the viewer of the television device consists only of a child attribute that is an attribute of children under a first predetermined age;
When the attribute configuration estimated by the attribute configuration estimating unit is the first attribute configuration, the attribute configuration correction unit adjusts the attribute configuration of the viewer of the television device to the child attribute and the attribute configuration other than the child attribute. 2, the attribute configuration is corrected to include attribute A, which is an attribute having the highest value of the attribute-specific existence probability specified by the attribute configuration estimation unit, among the attributes satisfying the condition of being at least a predetermined age of 2. A viewing information processing device. In the viewing information processing apparatus according to claim 4,
The attribute configuration correcting unit, when the attribute configuration estimated by the attribute configuration estimating unit is the first attribute configuration, excludes the child attribute and the attribute satisfying the second predetermined age or older condition. Among the attributes, the value of the attribute-specific existence probability specified by the attribute composition estimation unit is the same value as Pa, which is the value of the attribute-specific existence probability related to the attribute A, or a value higher than Pa. A viewing information processing apparatus, wherein, when an attribute exists, the viewing information processing apparatus is configured to correct the attribute configuration of the viewer of the television device so as to further include the existing attribute. The viewing information processing device according to claim 1,
the predetermined anomalous attribute configuration includes a second attribute configuration in which the number of attribute types included in the viewer attribute configuration of the television device is zero;
When the attribute configuration estimated by the attribute configuration estimating unit is the second attribute configuration, the attribute configuration correction unit sets the attribute configuration of the viewer of the television device to a third predetermined age or older. A viewing information processing apparatus, wherein the attribute configuration is corrected to include an attribute B, which is an attribute having the highest value of the attribute-specific existence probability specified by the attribute configuration estimating unit among the attributes to be satisfied. In the viewing information processing apparatus according to claim 6,
The attribute that satisfies the condition of being at least the third predetermined age is an attribute other than the child attribute that is the attribute of children under the first predetermined age,
When the attribute configuration estimated by the attribute configuration estimating unit is the second attribute configuration, the attribute configuration correcting unit determines that the value of the attribute-specific existence probability specified by the attribute configuration estimating unit regarding the child attribute is the above A fourth value that is the same as or higher than Pb, which is the value of the attribute-specific existence probability specified for attribute B, and that attribute B is higher than the third predetermined age The viewing information processing apparatus, wherein the viewing information processing apparatus is configured to correct an attribute configuration of the viewer of the television device so as to further include the child attribute when the attribute is a predetermined age or older. In the viewing information processing device according to any one of claims 1 to 7,
The predetermined abnormal attribute configuration includes a third attribute configuration in which the number of attribute types of the viewer of the television device is equal to or greater than a predetermined number,
When the attribute configuration estimated by the attribute configuration estimating unit is the third attribute configuration, the attribute configuration correcting unit adjusts the attribute configuration of the viewer of the television device from among the attributes included in the third attribute configuration. 7. A viewing information processing apparatus, wherein one or more attributes including at least an attribute having the lowest value of the existence probability for each attribute specified by the attribute configuration estimating unit are corrected to an attribute configuration in which one or more attributes are deleted. Output from a television device comprising a television broadcast receiver and a viewing data output device capable of outputting viewing log information indicating which channel or broadcasting station affiliated television broadcast was viewed through the receiver and when a first step of acquiring viewing log information;
A viewing information processing method comprising: a second step of estimating an attribute configuration of a viewer of the television device from the viewing log information acquired in the first step,
In the second step, when estimating the attribute configuration of the viewers of the television device, for each attribute type, an attribute-specific existence probability, which is a probability that a viewer of the attribute exists, is identified, and the identified attribute is estimating a predetermined attribute determined based on the separate existence probability as an attribute of a constituent member of an audience of the television device;
When the attribute configuration estimated in the second step is a predetermined abnormal attribute configuration, the attribute configuration of the viewer of the television device is determined according to the presence probability by attribute of the viewer of the television device, A viewing information processing method, further comprising a third step of correcting and determining from the abnormal attribute configuration.

Images