JP6951382B2 - Media contact status survey device and media contact status survey method - Google Patents

Description

The present invention relates to a media contact status investigation device and a media contact status investigation method, and in particular, the scale of the number of contacts to a certain media and the number of contacts to the media that distributes the same contents as the media on a communication network. Regarding the scale of the media contact status survey device and the media contact status survey method for investigating.

Information transmission media such as televisions, radios, newspapers and magazines are already well known. Further, in recent years, media such as IP (Internet Protocol) simul radio and digital newspapers, which distribute the same contents as certain media via a communication network such as the Internet, have become widespread.

In addition, the status of contact with each media, specifically the scale of the number of contacts (for example, TV audience rating, radio listening rate, newspaper and magazine subscription rate, etc.), is advertised using each media. It is important information for people (for example, advertising companies and advertising clients), and it is desirable that it be investigated regularly. Therefore, for example, in the case of TV ratings surveys, in general, a dedicated device is installed in the house of the surveyed household, and data indicating the viewing status is transmitted to the survey center at intervals of several minutes through the device. ing.

Further, in recent years, a technique for reducing an investigation burden such as an audience rating has been developed, and an example thereof is the device described in Patent Document 1. The audience rating investigation device described in Patent Document 1 is installed in a home and transmits data related to a broadcast program received at the home as audience rating investigation data. This device receives the extracted data whose amount of data is reduced by using a specific algorithm from the broadcast output data of each broadcasting station, accumulates the received extracted data, and displays the broadcast output data actually viewed by the broadcast receiver. By recording and collating the extracted data with the broadcast output data, the broadcasting station that was being viewed is determined, and the determined station is transmitted to the viewing rate survey center as viewed data. According to such a device, it is possible to reduce the burden of hardware that must be installed in the surveyed home and also reduce the burden of creating data to be transmitted to the surveyed home.

Japanese Unexamined Patent Publication No. 2001-298433

However, depending on the type of media, it is not costly to investigate the contact situation from the viewpoint of cost effectiveness, and therefore it is necessary to reduce the number of investigations or conduct the investigation by a relatively inexpensive method. In addition, surveys by inexpensive methods may not be conducted on the basis of statistical theory, which may affect the accuracy of the survey. Also, when requesting a survey to a randomly selected person, it may be difficult to obtain the consent of that person (for example, to meet that person and have them cooperate in the survey).
On the other hand, in the case of the above-mentioned online distribution type media, it is possible to relatively easily identify the scale of the number of contacts from log information and the like. Based on this situation, in order to conduct a contact status survey on media for which the number of surveys and methods are limited, using the contact status with other media that distributes the same content as that media via a communication network. Technology development is required.

The present invention has been made in view of the above circumstances, and an object of the present invention is to solve the following object. That is, the present invention solves the above-mentioned problems of the prior art and estimates the contact status with a certain medium by using the contact status with another medium that distributes the same content as the media on the communication network. It is an object of the present invention to provide a media contact situation investigation device and a media contact situation investigation method capable of capable.

In order to achieve the above object, the media contact status investigation device of the present invention includes a first data acquisition unit that acquires first data indicating the scale of the number of contacts to the first media at the time of the first investigation, and a first data acquisition unit. (Ii) To the second data acquisition unit that acquires the second data indicating the scale of the number of contacts to the second media that distributes the same contents as the first media at the time of the survey on the communication network, and to the first media. Estimate the scale of the number of contacts to the first media at the time of estimation and the correspondence acquisition unit that acquires the correspondence between the scale of the number of contacts and the scale of the number of contacts to the second media. It has an estimation unit, and the number of settings at the time of the second survey is larger than the number of settings at the time of the first survey within a unit period, and the estimation unit is the second set indicated by the second data. Of the scale of the number of contacts to the media, the scale of the number of contacts to the first media at the time of the estimation target based on the scale at the time of the second survey that matches the time of the estimation target and the correspondence relationship. Is characterized by estimating.

In the media contact status investigation device of the present invention configured as described above, the scale of the number of contacts to the first media and the contacts to the second media that distribute the same contents as the first media via the communication network. Identify the correspondence between the scale of numbers. In addition, the number of surveys on the scale of the number of contacts to the second media (specifically, at the time of the survey) is larger than the number of surveys on the scale of the number of contacts to the first media. Then, the scale of the number of contacts to the first media at the time of the estimation target is estimated from the above-mentioned correspondence and the scale of the number of contacts to the second media at the time of the estimation target. As a result, it is possible to conduct a contact status survey on the first media with a smaller number of surveys by using the contact status with the second media with a larger number of surveys.

Further, in the media contact situation investigation device, the correspondence acquisition unit performs machine learning using the first data and the second data acquired in the past, thereby performing machine learning at the time of the first investigation and the first investigation. (Ii) It is preferable to obtain a mathematical model showing the correspondence at the time points corresponding to both of the time points of the survey.
According to the above configuration, when specifying the correspondence relationship regarding the scale of the number of contacts of each of the first media and the second media, it is specified by machine learning using the first data and the second data acquired in the past. Therefore, it is possible to identify a more appropriate correspondence.

Further, in the media contact status investigation device, the correspondence acquisition unit is included in the scale of the number of contacts to the first media and the scale of the number of contacts to the second media at the time of the first investigation. The ratio of one to the other may be calculated as the correspondence.
According to the above configuration, it is possible to estimate the scale of the number of contacts to the first media by a simpler method.

Further, in the media contact situation investigation device, the correspondence acquisition unit is monthly from the scale of the number of contacts to the first media and the scale of the number of contacts to the second media at the time of the first investigation. It is preferable to calculate the ratio a plurality of times and calculate the ratio by averaging the monthly ratios calculated a plurality of times.
According to the above configuration, the ratio is calculated by averaging the monthly ratios, so that a more credible value can be obtained as a result of calculating the ratios.

Further, in the media contact status survey device, the first data acquisition unit determines whether or not each of the survey subjects selected according to the statistical method has contact with the first media at the time of the first survey. The first data is acquired by aggregating the survey results at the time of the survey, and the second data acquisition unit obtains the second data indicating the number of unique users of the second media at the time of the second survey. The second data to be acquired may be generated by collecting the log information generated at the time of accessing the second media at any time and analyzing the log information.
According to the above configuration, usually, the scale of the number of contacts to the first media is obtained based on the result of the survey targeting the survey subjects selected according to the statistical method. It is possible to collect log information that accesses the media at any time, specify the number of unique users of the second media, and estimate the scale of the number of contacts to the first media from the specified number of unique users.

Further, in the media contact situation investigation device, the first media is a broadcasting medium, and the estimation unit broadcasts the scale of the number of contacts to the first media at the time of the estimation target. Estimated by station, it is suitable.
According to the above configuration, it is possible to grasp the estimation result regarding the scale of the number of contacts to the first media for each broadcasting station.

Further, in the media contact situation investigation device, the estimation unit uses the scale at the time of the estimation target, the day of the week at the time of the estimation target, among the scales of the number of contacts to the second media indicated by the second data. It is more preferable to estimate the scale of the number of contacts to the first media at the time of the estimation target time based on the correspondence.
According to the above configuration, when estimating the scale of the number of contacts to the first media at the time of estimation, it is possible to estimate by considering the day of the week at the time of estimation. As a result, a more reasonable estimation result can be obtained regarding the scale of the number of contacts to the first media.

Further, in the media contact status investigation device, the first data acquisition unit indicates the scale of the number of contacts to the first media at the time of the first investigation for each attribute of the contacts to the first media. It is more preferable that the first data is acquired and the estimation unit estimates the scale of the number of contacts to the first media at the time of the estimation target for each attribute of the contacts to the first media. ..
According to the above configuration, it is possible to grasp the estimation result about the scale of the number of contacts to the first media for each attribute of the contacts to the first media.

Further, in the media contact situation investigation device, the estimation target time point may be different from the first investigation time point and coincide with any of the second investigation time points.
According to the above configuration, contact with the first media at a time when the scale of the number of contacts to the first media cannot be measured because the first data cannot be obtained (that is, a time different from the time of the first survey). Estimate the scale of the number of people. In such a case, the function of the media contact status investigation device of the present invention will be more meaningfully exhibited.

Further, in the media contact status survey device, the first data acquisition unit obtains the first data based on the presence or absence of contact of the first media at the time of the first survey entered in the entry form by the survey subject. Once obtained, it is suitable.
According to the above configuration, the survey subject fills in the entry form with or without contact with the first media, and the first data is acquired based on the entry result. In this case, it is difficult to increase the number of surveys because it takes a relatively long time to survey the scale of the number of contacts with the first media, and a new survey target (person who responds to the survey request) is secured. It is also difficult to do. In such a situation, the effect of the present invention of estimating the scale of the number of contacts to the first media by using the scale of the number of contacts to the second media becomes more meaningful.

Further, the above-mentioned problem is the first data acquisition step in which the computer acquires the first data indicating the scale of the number of contacts to the first media at the time of the first investigation according to the media contact situation investigation method of the present invention. And the second data acquisition process in which the computer acquires the second data indicating the scale of the number of contacts to the second media that distributes the same contents as the first media at the time of the second survey on the communication network. Correspondence between the scale of the number of contacts with the first media and the scale of the number of contacts with the second media at the time when the computer corresponds to both the time of the first survey and the time of the second survey. The computer has a correspondence acquisition step of acquiring the first data and the second data based on the first data, and an estimation step of estimating the scale of the number of contacts to the first media at the time of estimation target time. In the unit period, the number of settings at the time of the second survey is larger than the number of settings at the time of the first survey, and in the estimation step, the computer contacts the second media indicated by the second data. Of the scale of the number of persons, the scale of the number of contacts to the first media at the time of the estimation target is estimated based on the scale at the time of the second survey that matches the time of the estimation target and the correspondence relationship. It is solved by.
With the above method, it is possible to estimate the contact status with the first media by using the contact status with the second media that distributes the same contents as the first media on the communication network.

According to the present invention, the contact status with the first media is estimated by using the contact status with the second media that distributes the same contents as the first media via the communication network. This makes it possible to estimate the contact status with the first media, which has a smaller number of surveys, by using the contact status with the second media, which has a larger number of surveys. As a result, it becomes possible to grasp the contact situation with the first media in more detail without increasing the number of investigations.

It is explanatory drawing about the media contact situation investigation. It is a figure which shows the output example about the estimation result of a radio listening rate (the 1). It is a figure which shows the output example about the estimation result of a radio listening rate (the 2). It is a figure which shows the output example about the estimation result of a radio listening rate (the 3). It is a figure which shows the output example about the estimation result of a radio listening rate (the 4). This is a screen example for displaying the estimation result of the radio listening rate for each attribute of the radio listener. It is a figure which shows the estimation result of the radio listening rate for the radio listener of a certain sex and a certain age (the 1). It is a figure which shows the estimation result of the radio listening rate for the radio listener of a certain sex and a certain age (the 2). It is a figure which shows the estimation result of the radio listening rate for the radio listener of a certain sex and a certain age (the 3). It is a block diagram which shows the structure of the media contact situation investigation apparatus. It is a figure which shows the data processing flow which concerns on a media contact situation investigation. It is a graph which shows the investigation result of the radio listening rate and the IP simul radio listening rate. It is explanatory drawing of the procedure of estimating the number of radio listeners by applying a ratio.

The media contact status investigation device and the media contact status investigation method according to one embodiment of the present invention (the present embodiment) will be described in detail below with reference to the preferred embodiments shown in the accompanying drawings.
It should be noted that the embodiments described below are merely examples for facilitating the understanding of the present invention, and do not limit the present invention. That is, the present invention may be modified or improved from the embodiments described below without departing from the spirit of the present invention. Also, of course, the present invention includes an equivalent thereof.
The daily cycle may be a regular cycle from 0:00 to 23:59, or a cycle adjusted to business hours from N: 0 to (23 + N): 59. It may be, for example, from 5 o'clock to 28:59 (N is a natural number).

<< About media contact situation survey >>
In explaining the media contact status investigation device and the media contact status investigation method according to the present embodiment, the media contact status investigation will be described with reference to FIG. FIG. 1 is an explanatory diagram of a media contact situation survey. For convenience of illustration, in FIG. 1, the number of survey subjects and users, which will be described later, is smaller than the actual number.

The media contact situation survey is conducted to grasp the scale of the number of contacts for media such as radio, television, newspapers and magazines. Here, the scale of the number of contacts corresponds to the number of media contacts (absolute number) or the ratio of contacts to the population (for example, audience rating, listening rate, subscription rate, etc.). Further, in the present embodiment, a media contact situation survey for grasping the scale of the number of contacts to the broadcasting media, particularly the scale of the number of contacts to the radio is carried out.

There are two types of radio: terrestrial radio (radio broadcasting) and IP (Internet Protocol) simul radio. IP simulcast is a service that simultaneously simulcasts radio broadcasts on the Internet (an example of a communication network), and can be used by communication devices such as personal computers, smartphones, and tablet terminals.

Then, in the media contact situation survey of the present embodiment, the scale of the number of contacts is investigated for each of the radio broadcasting and the IP simul radio. Here, the relationship between radio broadcasting and IP simul radio corresponds to the relationship between the first media and the second media. Explaining the relationship between the first media and the second media, both the first media and the second media are broadcasting media, and the second media has the same contents as the first media (contents such as programs, articles and advertisements). Etc.) are distributed via a communication network such as the Internet. That is, in the present embodiment, radio broadcasting corresponds to the first medium, and IP simul radio corresponds to the second medium. Then, in the present embodiment, as will be described later, the scale of the number of contacts is to be investigated for each of the first media and the second media by different methods. In the present embodiment, radio broadcasting corresponds to the first media and IP simul radio corresponds to the second media, but the present invention is not limited to this, and IP simul radio corresponds to the first media and the second media. It corresponds to both media, and the scale of the number of contacts to IP Simul Radio may be investigated by two types of investigation methods described later.

In the present embodiment, the media contact situation survey is to be conducted for the radio, but the contents described below are media for other types of media (for example, television, newspapers and magazines). It can also be applied to contact status surveys. Regarding the media contact situation survey targeting television, for example, television (television broadcasting) broadcast by terrestrial broadcasting, BS, CS, etc. is the first media, and programs, etc. (for example, BML tags) are set via the Internet. The television that delivers the content) can be the second medium. Further, regarding the media contact situation survey targeting newspapers and magazines, newspapers and magazines, which are paper media, can be used as the first media, and digital newspapers and electronic books can be used as the second media.

The survey targeting the first media, radio (terrestrial radio broadcasting in this embodiment), surveys the scale of the number of contacts at the time of the first survey, specifically, the listening rate at the time of the first survey. do. The first survey time point is the survey target date and time set for surveying the radio listening rate, for example, a predetermined survey period (for example, one week, more specifically, a specific week in an even month). ) Is set in 15-minute units on each day.

The setting at the time of the first survey is not particularly limited, but it is desirable to set it at an appropriate date and time in consideration of the cost of the radio listening rate survey. The survey time unit of the radio listening rate on each day of the survey period is not limited to 15 minutes, but may be 1 to several seconds, 1 to several minutes, 1 to several hours, or 1 It may be on a daily basis.
In addition to the real-time listening rate at the time of the first survey, the listening rate at the time of the first survey includes the case where programs and commercials broadcast at the time of the first survey are listened to within a certain period after broadcasting. The listening rate (so-called time-shifted listening rate) may be included.

Explaining the radio listening rate survey in detail, for example, it is carried out several times a year (specifically, once every two to six months). In the radio listening rate survey, research company R, which conducts media contact status surveys, targets thousands of survey subjects selected according to statistical methods (specifically, random strain sampling method). Then, the presence or absence of listening to the radio at the time of the first survey will be investigated. More specifically, a survey form (not shown) will be distributed to each survey subject, and each survey subject will be asked in a diary format whether or not they listen to the radio on each day during the survey period in 15-minute units. Fill out the above form. There are multiple radio broadcasting stations, which are the first media, and each survey subject fills in the presence or absence of radio listening for each broadcasting station. In addition, the above-mentioned entry form may be mailed in the form of a paper medium, or may be transmitted in an electronic data format to a smartphone, tablet terminal or personal computer (PC) owned by the survey subject.

After the end of the survey period, the survey company R collects the completed entry forms from each survey subject, and aggregates the survey results using the computer owned by the survey company R (hereinafter referred to as the survey computer 10). Calculate the radio listening rate from the aggregated results. As a result, data indicating the radio listening rate for each broadcasting station at the time of the first survey (specifically, the time of the survey set every 15 minutes on each day of the survey period) is acquired. This data corresponds to the "first data" of the present invention.

Further, in the present embodiment, the research company R grasps the age and gender as attributes of each survey subject in advance at the time of the survey, and aggregates each survey result for each attribute when totaling the survey results. .. As a result, as the first data, data showing the radio listening rate at the time of the first survey by attribute (gender and age) will be acquired.

The type and number of attributes used when investigating the listening rate of radio is not particularly limited, and attributes other than gender and age (for example, residential area, etc.) may be included. Further, the attributes are not limited to demographic attributes such as gender and age, and may be psychographic attributes such as behavior patterns, hobbies, tastes, and interests.

On the other hand, in the survey targeting IP Simul Radio, which is the second media, the scale of the number of contacts to the radio at the time of the second survey, specifically, the number of unique users who accessed IP Simul Radio at the time of the second survey. (Hereinafter, for convenience, the number of unique users of IP Simul Radio) will be investigated. The second survey time point is the survey target date and time set for surveying the number of unique users of IP Simul Radio, and is set in units of 15 minutes every day, for example. The setting at the time of the second survey is not particularly limited, but it is desirable to set it at an appropriate date and time. The survey time unit for the number of unique users of IP Simul Radio is not limited to 15 minutes, but may be 1 to several seconds, 1 to several minutes, 1 to several hours, or 1 day. But it may be. In addition, the survey of the number of unique users of IP Simul Radio is not limited to the case where it is conducted every day, and it may be conducted once every few days, or once every one to several weeks. It is also good (provided that the survey time is set more than the radio listening rate).

Explaining in detail the survey of the number of unique users of IP Simul Radio, the data provider V, which provides data to the survey company R, surveys the number of unique users of IP Simul Radio at the time of the second survey, and the data showing the survey results. Is generated and sent to the research company R. This data corresponds to the "second data" of the present invention.

More specifically, the data provider V sends the log information generated when each user (contact person to the IP simul radio) accesses the IP simul radio from the server for operating the IP simul radio at any time. collect. This log information includes data for specifying the user ID of the user and the usage time of the IP simul radio. Further, the log information further includes information for identifying a radio broadcasting station (hereinafter, simply a broadcasting station). Here, the broadcasting station indicated by the log information means a broadcasting station that is a source of a radio program or the like listened to by the user through IP Simul Radio (more specifically, a broadcasting station that broadcasts the program or the like by terrestrial wave). do.

Then, the data providing company V analyzes the collected log information using the computer owned by the data providing company V (hereinafter, the data providing computer 11). As a result, data indicating the number of unique users of the IP simul radio at the time of the second survey (specifically, the time of the survey set every 15 minutes every day) for each broadcasting station, that is, the second data is generated.

Further, in the present embodiment, the data providing company V associates the information registered for each user to use the IP simul radio, specifically, personal information such as the age and gender of each user with the user ID. I remember. Then, when the data provider V analyzes the log information and specifies the number of unique users of the IP simul radio, it also specifies the age and gender of each user. As a result, as the second data, data showing the number of unique users of the IP Simul Radio at the time of the second survey by attribute (gender and age) will be acquired.

The type and number of attributes used when investigating the number of unique users of IP Simul Radio is not particularly limited, and even if attributes other than gender and age (for example, residential area, etc.) are included. good. Further, the attributes are not limited to demographic attributes such as gender and age, and may be psychographic attributes such as behavior patterns, hobbies, tastes, and interests.
Here, it is desirable that the type of attribute used when investigating the number of unique users of IP simul radio is the same as the type of attribute used when investigating the listening rate of radio.

By the way, regarding the radio listening rate, the radio listening rate is generally low, which may limit the survey budget, etc., making it difficult to conduct surveys frequently, and dedicated equipment such as the TV audience rating. It is also difficult to install the survey in the home of the survey subject and conduct the survey. Therefore, the survey of the radio listening rate is usually conducted by distributing the entry form to the survey subjects and asking them to fill in the presence or absence of radio listening as described above, and the number of surveys is relatively small. For example, the survey is conducted once every few months.

On the other hand, regarding the number of unique users of IP Simul Radio, log information that is the source information is collected at any time, so it is possible to investigate frequently, and the number of investigations is relatively large, for example, 15 every day. The survey is conducted on a minute-by-minute basis. That is, within the unit period (specifically, one year), the number set at the time of the second survey is larger than the number set at the time of the first survey (in other words, the frequency of acquiring the second data is the first. It is set more than the data acquisition frequency).

Therefore, in the present embodiment, the research company R uses the research computer 10 to analyze the first data and the second data obtained by the research, and between the radio listening rate and the number of unique users of the IP simul radio. Clarify the correspondence. Then, the research company R uses the research computer 10 to estimate the radio listening rate at a certain time point (hereinafter referred to as the estimation target time point) based on the above correspondence and the number of unique users of the IP simul radio. As a result, an estimation result can be obtained for the radio listening rate with a small number of surveys, and for example, it is possible to grasp (know the estimated value) the radio listening rate at the date and time when the radio listening rate is not actually surveyed.

The research company R uses a predetermined platform to graph and visualize the estimation result of the radio listening rate as shown in FIGS. 2 to 5, and outputs the output contents of the advertising company, the advertising client, etc. Present to the customer. By confirming the estimation result of the radio listening rate, the customer considers, for example, a radio advertisement placement plan.

Incidentally, FIGS. 2 to 5 are diagrams showing an output example of the estimation result of the radio listening rate. FIG. 2 is a line graph showing the estimation results of the hourly radio listening rate on a certain day for each broadcasting station. FIG. 3 is a heat map showing the estimation results of the radio listening rate of a certain broadcasting station on an hourly basis on each day of the week. FIG. 4 is a line graph showing the estimation result of the hourly radio listening rate on each day of the week for a certain broadcasting station. FIG. 5 is a bar graph showing the average value of the estimation results of the radio listening rate in a certain period to the broadcasting station.

The output examples of FIGS. 2 to 5 are merely examples, and the method of outputting the estimation result of the radio listening rate (for example, the type of graph) is not particularly limited. For example, a program in which the estimation result of the radio listening rate of the broadcast program is described in each program column of the broadcast program table on a certain day may be output. Further, as shown in FIGS. 6A to 6D, the estimation result of the radio listening rate is shown by the attribute of the radio listener (contact person to the radio), specifically, by the category determined according to the age and gender. Alternatively, it is possible to output a table. FIG. 6A is a screen example for displaying the estimation result of the radio listening rate for each sex and age category of the radio listener. Explaining the screen of the figure in detail, candidates for the sex age category are displayed in the pull-down list PB displayed on the right side of the screen. When one candidate is selected from these, the radio listening rate is estimated for the radio listeners belonging to the selected sex age category. As a result, a graph showing the time change of the radio listening rate is displayed for each broadcasting station in the graph display area on the screen. Then, when the candidates selected in the pull-down list PB are switched, the estimated result (graph) of the displayed radio listening rate changes accordingly. For example, when the category of "men and women 12 to 39 years old" is selected, the figure is shown. The estimated result shown in the figure is displayed in 6B, and when the category of "men and women 12 to 49 years old" is selected, the estimated result shown in the figure is displayed in FIG. 6C. The illustrated estimation result is displayed.

<< Configuration of media contact status investigation device according to this embodiment >>
Next, the media contact status investigation device (hereinafter, media contact status investigation device 100) according to the present embodiment will be described with reference to FIG. 7. FIG. 7 is a diagram showing the configuration of the media contact status investigation device 100.

As shown in FIG. 5, the media contact status survey device 100 according to the present embodiment is composed of a survey computer 10 owned by the research company R and a data providing computer 11 owned by the data providing company V. There is. The investigation computer 10 and the data providing computer 11 are connected to each other so as to be able to communicate with each other via a network. Each of the research computer 10 and the data providing computer 11 is a general personal computer or server computer, and is a memory such as a CPU, ROM and RAM, a hard disk drive, a communication interface, an input device such as a keyboard and a mouse, and a display. It also has an output device such as a printer as a constituent device. Further, a computer program for exerting each function (specifically, for realizing each functional unit described later) is installed in each of the investigation computer 10 and the data providing computer 11.

The survey computer 10 acquires the first data indicating the radio listening rate at the time of the first survey, and acquires the second data indicating the number of unique users of the IP simul radio at the time of the second survey from the data providing computer 11. .. In addition, the survey computer 10 analyzes the first data and the second data to determine the radio listening rate and the number of unique users of the IP simul radio at both the first survey time point and the second survey time point. Get the correspondence between them. Further, the survey computer 10 acquires the second data indicating the number of unique users of the IP simul radio at the time of the estimation target from the data providing computer 11, and utilizes the second data and the above correspondence to be the estimation target. Estimate the radio listening rate at the time.

In the present embodiment, the media contact status investigation device 100 is composed of two computers (investigation computer 10 and data providing computer 11), but the present invention is not limited to this, and the media contact status investigation is not limited thereto. The device 100 may be composed of one computer, and for example, the function of the data providing computer 11 may be mounted on the investigation computer 10.

Here, the configurations of the investigation computer 10 and the data providing computer 11 will be described again from the functional aspect.
As shown in FIG. 7, the investigation computer 10 has a first data acquisition unit 21, a second data acquisition unit 22, a correspondence acquisition unit 23, an estimation unit 24, and an output unit 25. These functional units are realized by the cooperation of the hardware device of the research computer 10 and the computer program installed in the research computer 10.

As shown in FIG. 7, the data providing computer 11 has a log information collecting unit 31, a second data generating unit 32, and a second data transmitting unit 33. These functional units are realized by the cooperation of the hardware device of the data providing computer 11 and the computer program installed in the data providing computer 11.
Hereinafter, each functional unit of the research computer 10 and the data providing computer 11 will be described in detail.

[First data acquisition department]
The first data acquisition unit 21 acquires the first data indicating the radio listening rate (the scale of the number of contacts with the first media) at the time of the first survey. In the present embodiment, the first data is data showing the radio listening rate for each broadcasting station for each attribute of the radio listener (specifically, for each gender and age).

Further, in the present embodiment, the first data acquisition unit 21 acquires the first data based on the presence or absence of radio listening at the time of the first survey entered in the entry form by each survey subject. Specifically, as described above, the survey company R selects survey subjects according to a statistical method and distributes an entry form to each survey subject. The research company R collects the completed form from each survey subject after the radio listening rate survey is completed. On this completed entry form, the presence or absence of radio listening on the survey target day is entered for each broadcasting station at 15-minute intervals. Employees of the survey company R input the contents of the collected entry form for each survey subject (that is, whether or not each broadcasting station listens to the radio) into the survey computer 10. The first data acquisition unit 21 aggregates the survey results based on the contents of the input form and calculates the radio listening rate (the ratio of the number of radio listeners to the survey population). As a result, the first data is acquired.
Regarding the input of the entry form, the entry contents may be input using an input device such as a keyboard, and the entry contents are input by OCR (optical character reader) after the entry contents are captured as image data by a scanner or the like. You may input by reading.

In addition, personal information of each survey subject including age and gender is registered in the survey computer 10. The first data acquisition unit 21 sets the radio listening rate for each broadcasting station at the time of the first survey for each attribute of the radio listener based on the contents of the input form and the personal information of the registered survey subject. calculate. That is, in the present embodiment, the first data indicating the radio listening rate for each broadcasting station for each attribute of the radio listener is acquired.

As mentioned above, since the number of surveys of the radio listening rate is about several times a year, the number of acquisitions of the first data (in other words, the number set at the time of the first survey in one year) is also relatively high. It is decreasing.

[Second data acquisition department]
The second data acquisition unit 22 acquires the second data indicating the number of unique users of the IP simul radio (scale of the number of contacts to the second media) at the time of the second survey. In the present embodiment, the second data is data showing the number of unique users for each broadcasting station for each attribute of the user (specifically, for each gender and age).

Further, in the present embodiment, the second data is provided from the data providing computer 11, and more specifically, the second data is transmitted from the second data transmitting unit 33 through the network. The second data acquisition unit 22 receives the second data from the second data transmission unit 33 through the network. As a result, the second data is acquired.
As described above, the second data is generated by collecting the log information generated at the time of accessing the IP simul radio at any time and analyzing the log information. That is, in the present embodiment, the number of unique users of the IP Simul Radio is sequentially investigated. In this way, the number of surveys of the number of unique users of IP Simul Radio is much higher than the number of surveys of the radio survey rate, so the number of acquisitions of the second data is larger than the number of acquisitions of the first data, in other words. Then, the number set at the time of the second survey in one year is larger than the number set at the time of the first survey.

[Correspondence acquisition department]
The correspondence acquisition unit 23 has acquired the correspondence (correlation) between the radio listening rate and the number of unique users of IP Simul Radio at the time corresponding to both the first survey time and the second survey time. Acquired based on one data and second data. Specifically, the correspondence acquisition unit 23 uses machine learning using the first data and the second data acquired in the past, strictly speaking, the first data and the second data indicating the survey results on the same date and time. By executing, a mathematical model showing the above correspondence is acquired. Here, as the machine learning technique, for example, a random forest can be used, but other techniques may be used, and specific examples thereof include a neural network, a support vector machine, bagging, and boosting. .. Further, the method for identifying the correspondence is not limited to machine learning, and for example, general linear regression analysis may be used.

Further, the first data and the second data to be used for machine learning are preferably data for the past one year (more specifically, the latest one year). In this case, it is possible to identify the correspondence between the two while grasping the changes in the radio listening rate and the number of unique users of the IP simul radio for one year. If the radio listening rate is estimated based on such a correspondence, a reasonable estimation result can be obtained for each period of the year. However, the present invention is not limited to this, and machine learning may be performed using data for the past one month, several months, or several weeks.

[Estimator]
Of the number of unique users of the IP simul radio indicated by the acquired second data, the estimation unit 24 determines the number of unique users at the time of the second survey that matches the estimation target time point, and the correspondence relationship acquired by the correspondence relationship acquisition unit 23. Based on this, the radio listening rate at the time of estimation is estimated. Specifically, the estimation unit 24 estimates by inputting various parameters including the date and time of the estimation target time and the number of unique users of the IP simul radio at the estimation target time into the mathematical model constructed by machine learning. Estimate the radio listening rate at the target time point.

Here, as the input parameters required when estimating the radio listening rate, the date and time and the day of the week corresponding to the estimation target time point, and the number of unique users of the IP simul radio at the estimation target time point (the acquired second data indicates). Value). That is, the estimation unit 24 estimates the radio listening rate at the estimation target time by inputting the number of unique users of the IP simul radio at the estimation target time and the date and time and the day of the week at the estimation target time into the above mathematical model.

The above three parameters are desirable as input parameters when estimating the radio listening rate, but parameters other than the above three parameters (for example, temperature, precipitation, snowfall and weather at the time of estimation, or listening location). Etc.) may be included. However, as a matter of course, it is desirable to set the type and number of input parameters so as to be the optimum type and number from the viewpoint of improving the estimation accuracy of the radio listening rate.

Further, the estimation target time point can be set to any time point as long as the second data about the second survey time point that matches the estimation target time point is acquired. For example, the estimated target time point may be a time point different from the first survey time point and coincide with any second survey time point. When such an estimation target time point is set, the radio listening rate is estimated at the date and time when the survey of the radio listening rate (strictly speaking, the survey requested to the survey target person) is not actually performed. Become. In such a case, since the uninvestigated radio listening rate can be estimated, the function of the media contact status investigation device 100 according to the present embodiment will be more meaningfully exhibited.

Further, in the present embodiment, the estimation target time point is set in units of 15 minutes. This makes it possible to stably estimate the radio listening rate over the entire day of the year (365 days). However, the setting unit at the time of estimation target time is not limited to 15 minutes, but may be any time unit, for example, 1 second to several seconds, 1 minute, 1 hour, or 1 day. You may.

Further, when the number of unique users of IP Simul Radio at the time of estimation target is input, when the number of unique users for each broadcasting station is input, the estimation unit 24 sets the radio listening rate at the time of estimation target for each broadcasting station (that is,). , By broadcasting station of the first media) can be estimated. Further, in the present embodiment, the estimation unit 24 can estimate the radio listening rate at the time of estimation target for each attribute of the radio listener (specifically, a category determined based on age and gender). ..

[Output section]
The output unit 25 outputs the radio listening rate estimated by the estimation unit 24 by the above-mentioned output method (for example, the output method shown in FIGS. 2 to 5). The content output by the output unit 25 can be displayed on a display or printed by a printer.
In the present embodiment, the estimation unit 24 estimates the radio listening rate in units of 15 minutes as described above, but the output unit 25 outputs the estimation result of the radio listening rate in units of 1 hour. The estimation result of the radio listening rate in units of 15 minutes is averaged and output every hour.

[Log Information Collection Department]
The log information collecting unit 31 collects the above-mentioned log information from the server of each broadcasting station of the IP simul radio through the network. The log information may be collected at any time, at a predetermined time on each day, or log information for several days to several weeks may be collected regularly.

[Second data generator]
The second data generation unit 32 is an IP simul radio in units of 15 minutes on each day based on the collected log information and the registration information regarding the age, gender, etc. of each user stored in the data providing computer 11. Data indicating the number of unique users of, that is, the second data is generated. In the present embodiment, the second data generation unit 32 generates second data as the second data, which indicates the number of unique users for each broadcasting station for each attribute of the user.

[Second data transmitter]
The second data transmission unit 33 transmits the second data generated by the second data generation unit 32 to the investigation computer 10 through the network. On the investigation computer 10, as described above, the second data acquisition unit 22 receives the second data.
In the present embodiment, the second data is transmitted at intervals of 15 minutes every day, but the present invention is not limited to this, and the second data may be transmitted every hour, every few hours, or every day. Second data for several weeks from the day may be sent together on a regular basis.

<< About media contact status survey method >>
Next, the media contact situation investigation method of the present invention will be described with reference to FIG. FIG. 8 is a diagram showing a data processing flow related to the media contact situation survey.
The media contact status investigation method of the present invention is adopted in the data processing flow shown in FIG. That is, each step in the data processing flow shown in FIG. 8 corresponds to a component of the media contact status investigation method of the present invention.

The data processing flow illustrated in FIG. 8 includes a first data acquisition process S001, a second data acquisition process S002, a correspondence acquisition process S003, an estimation process S004, and an output process S005. These steps are performed by the computers (investigation computer 10 and data providing computer 11) that constitute the media contact status investigation device 100, strictly speaking, the media contact status investigation device 100. Hereinafter, each step will be described.

[First data acquisition process]
In the first data acquisition step S001, the investigation computer 10 acquires the first data indicating the scale of the number of contacts to the first media at the time of the first investigation. More specifically, as described above, the survey company R collects the completed entry form from the survey subjects, and the employees of the survey company R, etc., have the presence or absence of radio listening during the survey period described in the entry form. Is input to the survey computer 10. After that, the survey computer 10 calculates the radio listening rate on each day of the survey period based on the contents of the input form. As a result, the first data is acquired.

In the present embodiment, since the entry form describes the presence or absence of radio listening for each broadcasting station on each day of the survey period (for example, one month) in units of 15 minutes, the radio listening rate is also for each broadcasting station. Calculated in 15 minute increments. Further, personal information (gender, age, etc.) of each survey subject is registered in the survey computer 10. When calculating the radio listening rate, the survey computer 10 calculates the radio listening rate by dividing it into the attributes (specifically, categories determined according to gender and age) of the survey subjects who listened to the radio. As described above, in the first data acquisition step S001, the first data indicating the radio listening rate for each broadcasting station on each day of the survey period in units of 15 minutes and for each attribute of the radio listener is acquired. ..

Further, in the present embodiment, the investigation of the radio listening rate is carried out at a relatively low frequency (for example, about several times a year), and the first data acquisition step S001 is carried out at the same frequency. Then, every time the survey of the radio listening rate is carried out, the survey computer 10 carries out the first data acquisition step S001 to acquire the first data indicating the result of each survey.

[Second data acquisition process]
In the second data acquisition step S002, the survey computer 10 and the data providing computer 11 cooperate to acquire the second data indicating the number of unique users of the IP simul radio at the time of the second survey. More specifically, as described above, the data providing computer 11 collects the log information of each user to the IP simul radio every 15 minutes every day, and analyzes the log information at each time point. The data providing computer 11 generates data indicating the number of unique users of the IP simul radio at each time point obtained from the analysis result, that is, second data, and transmits the generated second data to the investigation computer 10. The investigation computer 10 acquires the second data by receiving the second data from the data providing computer 11 through the network N.

In the present embodiment, the data providing computer 11 identifies the number of unique users of the IP simul radio for each broadcasting station by analyzing the log information for each broadcasting station. Further, personal information (gender, age, etc.) of each user is registered in the data providing computer 11. When specifying the number of unique users, the data providing computer 11 specifies the number of unique users by classifying the attributes of the users (specifically, categories determined according to gender and age). As described above, in the second data acquisition step S002, the second data indicating the number of unique users for each broadcasting station in units of 15 minutes every day and for each attribute of the user is acquired.

As described above, in this embodiment, the number of unique users of IP Simul Radio is investigated every 15 minutes every day. That is, the time point of the survey of the number of unique users of IP Simul Radio (at the time of the second survey) in one year (unit period) is set to be larger than the time point of the survey of the radio listening rate (at the time of the first survey). Then, the survey computer 10 and the data providing computer 11 carry out the second data acquisition step S002 at any time to acquire the number of unique users of the daily (365 days) IP simul radio.

[Correspondence acquisition process]
In the correspondence relationship acquisition step S003, the survey computer 10 has previously determined the correspondence relationship between the radio listening rate and the number of unique users of the IP simul radio at both the time points of the first survey and the time point of the second survey. Acquire based on the acquired first data and second data.
More specifically, when a sufficient amount of the first data and the second data for acquiring the correspondence relationship, specifically, the amount for the last one year is acquired, the investigation computer 10 performs the correspondence relationship acquisition process. Carry out S003. In the correspondence acquisition step S003, the survey computer 10 carries out machine learning (specifically, random forest) using the first data and the second data for the past year. As a result, a mathematical model representing the correspondence (correlation) between the radio listening rate and the number of unique users of IP simul radio is acquired.

[Estimation process]
In the estimation step S004, the survey computer 10 estimates the radio listening rate at the time of estimation. Specifically, the survey computer 10 uses the mathematical model acquired in the correspondence acquisition step S003 to perform a second survey that matches the estimation target time point among the number of unique users of the IP simul radio indicated by the acquired second data. Enter the number of unique users at the time point, and the date and time and day of the week at the time of estimation. As a result, the radio listening rate at the time of estimation is estimated.

Further, in the estimation step S004, when inputting the date and time of the estimation target time point in the mathematical model, if a date and time different from the first survey time point and coincide with any second survey time point is input, the radio listening rate will be increased. Radio listening rates can be estimated for dates and times when the survey is not actually conducted. Further, if the number of unique users of IP Simul Radio at the time of estimation target is input for each broadcasting station, the radio listening rate at the time of estimation target can be estimated for each broadcasting station. In addition, the estimation result of the radio listening rate can be estimated for each attribute of the radio listener (specifically, a category determined based on age and gender).

[Output process]
In the output step S005, the investigation computer 10 outputs the radio listening rate at the time of the estimation target estimated in the estimation step S004 by a predetermined output method (for example, the output method shown in FIGS. 2 to 5).
The output step S005 is performed when a sufficient amount (for example, one day's worth, several days' worth, one week's worth, or one month's worth) is obtained to present the estimation result of the radio listening rate to the customer. NS.

According to the data processing flow related to the media contact situation survey explained above, the number of surveys can be reduced because the survey is not costly, and the radio is surveyed by a simple and inexpensive survey method using an entry form. A stable and sufficient amount of estimation results can be obtained for the listening rate. Then, when considering the radio advertisement placement plan, the customer efficiently considers by referring to the output result of the radio listening rate provided by the research company R (for example, the output result shown in FIGS. 2 to 5). It becomes possible.

<< About the modified example >>
In the above-described embodiment, the correspondence between the scale of the number of contacts to the radio, which is the first media, and the scale of the number of contacts to the IP simul radio, which is the second media, was acquired in the past. It was decided to carry out machine learning using the first data and the second data and acquire a mathematical model as the above correspondence. Then, by inputting various parameters (date and time of the estimation target time, etc.) into the acquired mathematical model, it was decided to estimate the radio listening rate at the estimation target time. The above mathematical model accurately represents the correspondence (correlation) between the radio listening rate and the number of unique users of IP Simul Radio, and is effective in estimating the radio listening rate accurately.

On the other hand, as the correspondence relationship between the scale of the number of contacts to the radio, which is the first media, and the scale of the number of contacts to the IP simul radio, which is the second media, other than the above mathematical model can be considered. .. In the following, an example (modification example) of acquiring a correspondence relationship other than the mathematical model will be described with reference to FIGS. 9 and 10. 9 and 10 are explanatory views of a modified example, and FIG. 9 is a graph showing the survey results of the radio listening rate and the IP simul radio listening rate. The vertical axis in the figure indicates the listening rate (%). FIG. 10 is an explanatory diagram of a procedure for estimating the number of radio listeners by applying the ratio described later. The vertical axis in the figure indicates the number of unique users.

In the following, a specific example is the case where the radio listening rate survey is conducted once every two months, and strictly speaking, it is conducted in February, April, June, August, October, and December. It will be explained by listing in. However, it is not limited to this, and even in the modified example, it is possible to arbitrarily set the survey frequency and the survey implementation date of the radio listening rate (however, from the survey frequency of the number of IP simul radio users). Only when there are few).

In the modified example, as in the above embodiment, the radio listening rate is investigated as the scale of the number of contacts to the radio, and the number of unique users of the IP simul radio is investigated as the scale of the number of contacts to the IP simul radio. That is, also in the modified example, the first data indicating the radio listening rate at the time of the first survey is acquired, and the second data indicating the number of unique users of the IP simul radio at the time of the second survey is acquired. Furthermore, in the modified example, the listening rate (contact rate) of the IP simul radio is to be investigated at the same time as the investigation of the radio listening rate. That is, in the modified example, data indicating the listening rate of the IP simul radio at the time of the first survey (hereinafter referred to as the third data) is acquired. The third data is data showing the scale of the number of contacts with the second media at the time of the first survey, and is different from the second data.

Then, in the modified example, the ratio to one of the two is calculated as the correspondence relationship between the scale of the number of contacts to the radio and the scale of the number of contacts to the IP simul radio at the time of the first survey. .. Specifically, the ratio of the listening rate of IP simul radio at the same survey time to the radio listening rate at a certain survey time is calculated.

More specifically, in the modified example, the correspondence acquisition unit 23 calculates the above ratio based on the first data and the third data for the past one year. Specifically, the correspondence acquisition unit 23 calculates the radio listening rate (scale of the number of contacts to the radio) on each survey date from the first data for the past year in units of one hour. When the radio listening rate indicated by the first data is, for example, a listening rate in units of 15 minutes, the listening rate in units of 15 minutes may be averaged every hour and converted into a radio listening rate in units of 1 hour. ..

In addition, the correspondence acquisition department 23 has obtained the IP Simul Radio listening rate (the scale of the number of contacts with the IP Simul Radio) on the same day as the radio listening rate survey implementation date from the third data for the past year. Is calculated on an hourly basis. If the listening rate of the IP simul radio shown by the third data is, for example, the listening rate of 15 minutes, the listening rate of 15 minutes should be averaged every hour and converted into the listening rate of 1 hour. Just do it.

Through the above procedure, as shown in FIG. 9, the listening rates of the radio broadcast and the IP simul radio at the same survey time point (first survey time point) are calculated in 1-hour units. The radio listening rate and the IP simul radio listening rate are calculated for each broadcasting station.

After that, the correspondence acquisition unit 23 calculates the ratio (= radio listening rate / IP simul radio listening rate) in 1-hour units. Explaining the case shown in FIG. 9 as an example, for example, the ratio at 18:00 is calculated from the radio listening rate and the IP simul radio listening rate at the same time, and the calculated value is 4.56 (= 4. 1 / 0.9).
The ratio of each time is also calculated for each broadcasting station.

Further, the correspondence acquisition unit 23 repeats the above procedure a plurality of times by changing the implementation month, and specifically, 6 times (that is, February, April, June, August, 10). Monthly and December) Repeat. As a result, the monthly ratio will be calculated for June on an hourly basis. Then, the correspondence acquisition unit 23 averages (additionally averages) the ratios of the same time among the monthly ratios calculated a plurality of times (six times). This averaged hourly ratio corresponds to the correspondence between the size of the number of contacts to the radio and the size of the number of contacts to the IP simul radio in the modified example. In addition, by averaging the monthly ratios calculated multiple times to calculate the ratio, the influence of seasonal fluctuations can be taken into consideration, and as a result of calculating the ratio, a more credible value can be obtained. ..
The average ratio of monthly ratios is also calculated for each broadcasting station.

After the above ratio is calculated, the estimation unit 24 estimates the number of radio listeners at the time of estimation target for each broadcasting station based on the above ratio. More specifically, the estimation unit 24 describes the ratio of the calculated ratio to the number of unique users at the time of estimation target among the number of unique users of the IP simul radio indicated by the acquired second data, which corresponds to the time at the time of estimation target. Multiply. This makes it possible to estimate the number of radio listeners at the time of estimation. To explain with a specific example, it is assumed that the number of IP simul radio users on a certain day changes for a certain broadcasting station A as shown in the graph shown in FIG. When the estimation unit 24 estimates the number of radio listeners of the broadcasting station A at 19:00 on a certain day, the number of unique users of the IP simul radio at the same time (indicated by the symbol P in FIG. 10) is set to the broadcasting station A. Multiply the 19:00 ratio α calculated for the station. The number of people (= α × P) calculated in this way corresponds to the estimation result of the number of radio listeners of broadcasting station A at 19:00 on a certain day.

As described above, the modified example differs from the above-described embodiment in that the ratio of the two is calculated as the correspondence relationship between the scale of the number of contacts to the radio and the scale of the number of contacts to the IP simul radio. However, in other respects, it is common to the above-described embodiment, and it is possible to obtain the same effect as that of the above-mentioned embodiment.

<< Other Embodiments >>
The media contact status investigation device and the media contact status investigation method of the present invention have been described above with an example, but the above-described embodiment is merely an example, and other examples are also conceivable.
For example, in the above-mentioned embodiment, when conducting a survey of the radio listening rate, it was decided to distribute an entry form to the survey subjects and have them fill in the entry form with or without radio listening on each day of the survey period. However, the present invention is not limited to this, and similar to the TV audience rating, a device that detects the presence or absence of radio listening and transmits a detection signal to the center is installed in the survey subject's house, and the radio is used using the device. A listening rate survey may be conducted. However, in the above-described embodiment, it is difficult to increase the number of investigations because it takes time and effort to investigate the radio listening rate. Under such circumstances, the effect of the present invention of estimating the radio listening rate by utilizing the number of unique users of IP Simul Radio becomes even more meaningful.

Further, in the above embodiment, the estimated target time point at which the radio listening rate is estimated is set to a date and time different from the time point at which the radio listening rate is investigated (that is, the time point at which the first survey is conducted). This makes it possible to obtain an estimation result of the radio listening rate for the date and time when the radio listening rate survey is not carried out in the first place. However, the present invention is not limited to this, and a date and time that coincides with the time of the first survey may be set as the time of estimation. That is, the estimation result of the radio listening rate may be acquired for the date and time when the investigation of the radio listening rate is carried out. In this case, since both the survey result (actual measurement result) and the estimation result can be obtained for the radio listening rate, it is possible to compare the two results and verify the estimation accuracy.

10 Investigation computer 11 Data provision computer 21 First data acquisition unit 22 Second data acquisition unit 23 Correspondence relationship acquisition unit 24 Estimating unit 25 Output unit 31 Log information collection unit 32 Second data generation unit 33 Second data transmission unit 100 Media contact status survey device PB pull-down list R Survey company V Data provider

Claims (11)

The first data acquisition department that acquires the first data indicating the scale of the number of contacts to the first media at the time of the first survey,
The second data acquisition unit that acquires the second data indicating the scale of the number of contacts to the second media that distributes the same contents as the first media at the time of the second survey on the communication network.
The correspondence acquisition unit that acquires the correspondence between the scale of the number of contacts to the first media and the scale of the number of contacts to the second media, and
It has an estimation unit that estimates the scale of the number of contacts to the first media at the time of estimation target, and has.
Within the unit period, the number set at the time of the second survey is larger than the number set at the time of the first survey.
The estimation unit estimates the number of contacts to the second media indicated by the second data based on the scale at the time of the second survey that matches the time of the estimation target and the correspondence relationship. A media contact situation survey device for estimating the scale of the number of contacts to the first media at a target time point. By performing machine learning using the first data and the second data acquired in the past, the correspondence acquisition unit performs the machine learning at a time corresponding to both the first survey time point and the second survey time point. The media contact situation investigation device according to claim 1, which acquires a mathematical model showing a correspondence relationship. The correspondence acquisition unit determines the ratio of one of the scale of the number of contacts to the first media and the scale of the number of contacts to the second media at the time of the first survey to the other. The media contact situation investigation device according to claim 1, which is calculated as. The correspondence acquisition unit calculated the monthly ratio a plurality of times from the scale of the number of contacts to the first media and the scale of the number of contacts to the second media at the time of the first survey, and calculated the ratio a plurality of times. The media contact status investigation device according to claim 3, wherein the ratio is calculated by averaging the monthly ratios. The first data acquisition unit aggregates the survey results when investigating the presence or absence of contact with the first media at the time of the first survey for each of the survey subjects selected according to the statistical method. Then, the first data is acquired,
The second data acquisition unit acquires the second data indicating the number of unique users of the second media at the time of the second survey.
The media contact according to any one of claims 1 to 4, wherein the second data is generated by collecting log information generated at the time of accessing the second media at any time and analyzing the log information. Situation survey equipment. The first media is a broadcasting medium,
The media contact status investigation device according to any one of claims 1 to 5, wherein the estimation unit estimates the scale of the number of contacts to the first media at the time of the estimation target for each broadcasting station of the first media. .. Among the scales of the number of contacts to the second media indicated by the second data, the estimation unit is the estimation target based on the scale at the time of the estimation target, the day of the estimation target time, and the correspondence relationship. The media contact status investigation device according to any one of claims 1 to 6, which estimates the scale of the number of contacts to the first media at a time point. The first data acquisition unit acquires the first data indicating the scale of the number of contacts to the first media at the time of the first survey for each attribute of the contacts to the first media.
The estimation unit according to any one of claims 1 to 7, which estimates the scale of the number of contacts to the first media at the time of the estimation target for each attribute of the contacts to the first media. Media contact status survey device. The media contact situation investigation device according to any one of claims 1 to 8, wherein the estimation target time point is different from the first investigation time point and coincides with any one of the second investigation time points. The first data acquisition unit according to claims 1 to 9 for acquiring the first data based on the presence or absence of contact with the first media at the time of the first investigation entered in the entry form by the survey subject. Media contact status survey device. The first data acquisition process in which the computer acquires the first data indicating the scale of the number of contacts with the first media at the time of the first survey.
A second data acquisition process in which the computer acquires the second data indicating the scale of the number of contacts to the second media that distributes the same contents as the first media on the communication network at the time of the second survey.
A correspondence relationship acquisition process in which the computer acquires a correspondence relationship between the scale of the number of contacts to the first media and the scale of the number of contacts to the second media.
The computer has an estimation step of estimating the scale of the number of contacts to the first media at the time of estimation.
Within the unit period, the number set at the time of the second survey is larger than the number set at the time of the first survey.
In the estimation step, the computer is based on the scale of the number of contacts to the second media indicated by the second data, the scale at the time of the second survey that matches the time of the estimation target, and the correspondence relationship. , A method for investigating a media contact situation, which comprises estimating the scale of the number of contacts to the first media at the time of the estimation target.

Images