JP6730002B2 - Extraction device, extraction method, and extraction program - Google Patents

Description

The present invention relates to an extraction device, an extraction method and an extraction program.

In recent years, with the rapid spread of the Internet, advertisement distribution based on information acquired via a network has been actively performed. In such advertisement distribution, an advertisement value such as a probability that a distributed advertisement is clicked by a user is measured, and an index value indicating the measured advertisement effect may be used.

For example, as an example of the above-mentioned advertisement distribution, when there are a plurality of advertisement contents configured by combining a plurality of parts for one product, one of the index values indicating the advertising effect of each advertisement content. There is known a technique of enhancing the advertising effect of the advertising content by predicting the CTR (Click Through Rate) which is the above and delivering the advertising content using the predicted CTR.

JP, 2014-6684, A

However, with the above-described conventional technology, it has been difficult to distribute content with a high appealing effect to more users. Specifically, in the conventional technique, the target user group is narrowed down to deliver the advertising content, and thus the number of delivering the advertising content may decrease. That is, there is a possibility that it may not be possible to achieve both distribution of advertising content with a high advertising effect and achievement of the number of distributions required by the advertiser.

The present application has been made in view of the above, and an object of the present invention is to provide an extraction device, an extraction method, and an extraction program that can deliver content with a high appeal effect to more users.

The extraction device according to the present application includes a reception unit that receives a plurality of advertisement contents related to a common advertisement target, and a plurality of the advertisement contents for each user group classified based on the attributes of the users who are the distribution targets of the advertisement contents. Is distributed to the user from among the plurality of advertisement contents based on the correlation between the acquisition unit that acquires the index value indicating the advertising effect of the advertisement and the index group acquired by the acquisition unit and the user group. And an extraction unit that extracts the advertising content.

According to one aspect of the embodiment, it is possible to deliver content having a high appeal effect to more users.

FIG. 1 is a diagram illustrating an example of an extraction process according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the extraction processing system according to the embodiment. FIG. 3 is a diagram illustrating a configuration example of the advertising device according to the embodiment. FIG. 4 is a diagram illustrating an example of the advertisement information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of the user information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of the delivery log storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of the model storage unit according to the embodiment. FIG. 8 is a diagram showing an example of a data table according to the embodiment. FIG. 9 is a flowchart showing the extraction processing procedure (1) by the advertising device according to the embodiment. FIG. 10 is a flowchart showing the extraction processing procedure (2) by the advertising device according to the embodiment. FIG. 11 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the advertising device.

Hereinafter, modes (hereinafter, referred to as “embodiments”) for implementing an extraction device, an extraction method, and an extraction program according to the present application will be described in detail with reference to the drawings. Note that the extraction device, the extraction method, and the extraction program according to the present application are not limited by this embodiment. Further, the respective embodiments can be appropriately combined within a range in which the processing content is not inconsistent. Also, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicated description will be omitted.

[1. Overview of extraction process]
First, an example of the extraction process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of an extraction process according to the embodiment. FIG. 1 illustrates an example in which the advertising device 100 corresponding to the extraction device according to the present application performs a process of extracting advertising content distributed to a user.

The advertising device 100 is a server device that holds the advertising content, extracts the advertising content suitable for the user when receiving a request for advertising distribution from the user, and distributes the extracted advertising content. The advertising device 100 uses information about the user (hereinafter referred to as “user information”) to perform appropriate matching processing between the user who is the distribution target and the distributed advertising content. In other words, the advertising device 100 extracts advertising content that is expected to have a higher advertising effect by being delivered to a specific user. By using the advertising device 100, an advertiser (eg, a company that handles products and the like) who requests the distribution of advertisements is likely to be able to obtain some results from the user when the distribution of advertisements is used as an opportunity, or clicks. Realization of delivery of advertisement content that is expected to have a high rate, that is, advertisement content with a high advertising effect.

Here, the user information refers to user attribute information and the like. For example, the user information includes the user's age, sex, place of residence, preference, annual income, or the like. The user information specifies, for example, a cookie that is transmitted and received when the user terminal 10 (not shown in FIG. 1), which is an information processing terminal operated by the user, accesses a predetermined web server. As a result, it is acquired by the advertising device 100. It should be noted that the advertising device 100 is not limited to the above method, and may acquire user information by receiving information provided from a predetermined external device.

That is, in the example of FIG. 1, the advertising device 100 according to the embodiment acquires the user information and uses it for the extraction process of the advertising content. Below, the outline of the extraction processing of the advertising content performed by the advertising device 100 will be described along the flow.

As shown in FIG. 1, the advertiser CL01 prepares a plurality of advertising contents to be submitted to the advertising device 100 (step S11). Here, the plurality of prepared advertising contents are advertising contents that advertise one common target. That is, the advertiser CL01 can perform highly effective advertisement distribution to users having various attributes by preparing a plurality of different types of advertisement content for one advertisement target. Therefore, when preparing a plurality of advertisement contents, the advertiser CL01 assumes several user groups in which the users are classified based on the attributes of the users who are the distribution destinations. For example, the advertiser CL01 has a group of users whose age group is “20s” and their gender is “male”, a group of users whose age group is “40s” and their gender is “female”, etc. , Suppose some groups of users.

Then, the advertiser CL01 prepares the advertising content having the content expected to exert the advertising effect for each of the assumed user groups. Specifically, the advertiser CL01 prepares advertising contents that are expected to have a high appealing effect on "male in their 20s" and advertising contents that are expected to have a high appealing effect on "women in their 40s". To do. In the example of FIG. 1, it is assumed that the advertiser CL01 prepares the advertising content C01, the advertising content C02, the advertising content C03, and the advertising content C04 as a plurality of advertising content.

Then, the advertiser CL01 submits the advertising contents C01 to C04, which are a plurality of advertising contents related to the common advertising target, to the advertising device 100 (step S12). The advertising device 100 holds a plurality of advertising contents received from the advertiser CL01 as one advertising group.

Subsequently, the advertising device 100 receives a request for advertisement distribution from a user who operates the user terminal 10 that can browse a web page. For example, the advertising device 100 is a request for advertisement distribution transmitted from the user terminal 10 that has acquired the web page, and a request for distribution of advertisement content to be displayed in the advertisement frame provided on the acquired web page. To receive.

Here, the advertising device 100 randomly extracts advertisement content from the advertisement content C01 to C04 to the user who has transmitted the advertisement distribution request, and distributes the extracted advertisement content to the users U11 to U16 (step S13). That is, the advertising device 100 randomly extracts the delivered advertising content from the plurality of advertising content received from the advertiser CL01 until a predetermined stage. Specifically, the advertising device 100 randomly extracts the advertising content until a predetermined ratio (for example, 10% of the total distribution number) of the total distribution number of the advertising content received from the advertiser is achieved. To do. In the example illustrated in FIG. 1, the users U11 to U16 are users having different attributes. For example, the user U11 is a male and is in his twenties. The user U12 is a male and is a user in his or her 40s. The user U13 is a male and is in his 60s. Further, the user U14 is a woman who is in his or her twenties. The user U15 is a woman who is in his or her 40s. Further, the user U16 is a woman who is in his or her sixties. The user U11 includes the concept of a user group including not only the individual user U11 but also users having the same attributes as the user U11. The same applies to the other users U12 to U16.

Subsequently, the advertising device 100 acquires information regarding the advertising effect of the advertising content distributed to the user U11 (step S14). For example, the advertising device 100 uses the number of times the advertising content has been selected by the user U11 (for example, the number of clicks on the user terminal 10 operated by the user U11) or the distribution of the advertising content as a trigger to the advertiser CL01. The information about the profit brought to the advertiser CL01, such as the number of times the profit is brought, is acquired. Note that the advertising device 100 also acquires information regarding the advertising effect of the advertising content for the other users U12 to U16 as well. Then, the advertising device 100 calculates an index value indicating an advertising effect such as CTR based on the acquired information.

Then, the advertising device 100 optimizes the extraction process to be performed thereafter for the users U11 to U16 based on the acquired information regarding the advertising effect of the advertising content (step S15). As will be described later in detail, the advertising device 100 considers, among the plurality of advertising content submitted by the advertiser, which advertising content is targeted to which user group, which is expected to have the highest advertising effect. Therefore, the extraction process of the advertising content is optimized. For example, the advertising device 100 creates a prediction model for predicting the advertising effect based on the correlation between the index value indicating the advertising effect of the acquired advertising content and the user group. Then, the advertising device 100 uses the created prediction model to extract the advertising content predicted to have a high advertising effect.

After that, the advertising device 100 performs optimized extraction processing, extracts the advertising content distributed for each user, and distributes the extracted advertising content to the user (step S16). For example, the advertising device 100 can obtain the highest advertising effect by distributing the advertising content C03 among the advertising contents C01 to C04 to the user U11 who belongs to the user group of “male, twenties”. Suppose that it is judged. In this case, the advertising device 100 preferentially delivers the advertising content C03 to the user U11. Further, the advertising device 100 can obtain the highest advertising effect by distributing the advertising content C04 among the advertising contents C01 to C04 to the user U12 who is a user group of “male, 40s”. Suppose that it is judged. In this case, the advertising device 100 preferentially distributes the advertising content C04 to the user U12 who belongs to the user group “male, 40s”. As described above, the advertising device 100 extracts the optimized advertising content for each user based on the user information and the information such as the index value obtained by the advertising distribution, and distributes the extracted advertising content.

As described above, the advertising device 100 according to the embodiment receives submission of a plurality of advertising contents related to a common advertising target. Then, the advertising device 100 acquires an index value indicating the advertising effect of a plurality of advertising contents for each user group classified based on the attribute of the user who is the distribution target of the advertising contents. Then, the advertising device 100 extracts the advertising content to be delivered to the user from the plurality of advertising content based on the correlation between the acquired index value and the user group.

That is, the advertising device 100 according to the embodiment accepts submission of advertising content having a common advertising target and corresponding to a plurality of user groups. Specifically, the advertising device 100 is an advertising content that targets “automobile”, and advertising content that appeals “driving performance” and advertising content that appeals “load capacity” and “size of vehicle interior space”. And, we will accept submission of multiple advertising content such as advertising content that appeals for "utilization as entertainment such as driving". In this way, the advertiser can use the advertising device 100 to submit each advertising content that can be expected to appeal to various user groups even when advertising the same “car”. it can. Then, the advertising device 100 performs a process of extracting advertisement content that is expected to have a high appealing effect to each user who has been assigned the advertisement of “car”. At this time, the advertising device 100 improves the extraction accuracy of the advertising content by acquiring the advertising effect that is the result when each advertising content is distributed to each user layer and optimizing the extraction processing. As a result, the advertising device 100 can prevent a reduction in the number of distributions due to narrowing the target of advertising content to a specific user group while maintaining a high advertising effect. As a result, the advertising device 100 can deliver the content having a high appeal effect to more users.

Note that FIG. 1 illustrates an example in which the advertising device 100 acquires information regarding the advertising effectiveness of the actually distributed advertising content in step S14. Here, the advertising device 100 may acquire the information regarding the advertising effect of the advertising content by using a notification function realized by a web beacon or the like.

That is, the web page on which the advertising contents C01 to C04 are displayed may have a notification function regarding user information, which is realized by a web beacon or the like, embedded therein. For example, the web beacon has a function of allowing the user terminal 10 that has accessed the web page to access a transparent image or a very small image (also referred to as “clear GIF”) stored in the advertising device 100. Thereby, the advertising device 100 can receive and acquire the user information from the user terminal 10. For example, in the advertising device 100, the user clicks the advertising content on the user terminal 10 by the function realized by the web beacon, and the page linked to the advertising content (hereinafter referred to as “landing page” or “LP”). You may receive the information that you browsed. Further, the advertising device 100 may acquire information regarding the advertising effect on the user by receiving access information and the like from the user terminal 10 from a predetermined web server that provides the landing page.

[2. Configuration of extraction processing system]
Next, the configuration of the extraction processing system 1 including the advertising device 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram showing a configuration example of the extraction processing system 1 according to the embodiment. As illustrated in FIG. 2, the extraction processing system 1 according to the embodiment includes a user terminal 10, an advertiser terminal 20, a web server 30, and an advertising device 100. These various devices are connected via a network N so that they can communicate by wire or wirelessly. The extraction processing system 1 illustrated in FIG. 2 may include a plurality of user terminals 10, a plurality of advertiser terminals 20, and a plurality of web servers 30.

The user terminal 10 is an information processing device such as a desktop PC (Personal Computer), a notebook PC, a tablet terminal, a mobile phone, a PDA (Personal Digital Assistant), or the like. For example, the user terminal 10 acquires the web page from the website provided by the web server 30 by accessing the web server 30 according to the operation by the user. Then, the user terminal 10 displays the acquired web page on a display device (for example, a liquid crystal display).

The advertiser terminal 20 is an information processing device used by an advertiser who requests the advertisement device 100 to deliver an advertisement. The advertiser terminal 20 submits the advertising content to the advertising device 100 according to the operation by the advertiser. At this time, the advertiser terminal 20 submits a plurality of advertisement contents for one advertisement target in order to perform the advertisement distribution in which the advertising effect is expected to be high.

In some cases, the advertiser may request the agent to submit the advertisement content using the advertiser terminal 20 without submitting the advertisement content to the advertising device 100. In this case, it is the agency that submits the advertising content to the advertising device 100. In the following, the notation “advertiser” is a concept that includes not only the advertiser but also the agency, and the notation “advertiser terminal” includes not only the advertiser terminal but also the agency device used by the agency. Shall be a concept.

The web server 30 is a server device that provides various web pages when accessed from the user terminal 10. The web server 30 provides various web pages regarding, for example, a news site, a weather forecast site, a shopping site, a finance (stock price) site, a route search site, a map providing site, a travel site, a restaurant introduction site, and a web blog.

The web page distributed by the web server 30 includes an advertising space for displaying advertising content. Then, the web page including the advertisement frame includes an advertisement acquisition command for acquiring the advertisement content displayed in the advertisement frame. For example, in an HTML (HyperText Markup Language) file forming a web page, the URL of the advertising device 100 is described as an advertisement acquisition command. The user terminal 10 that has acquired the web page accesses the URL described in the HTML file or the like to receive the distribution of the advertising content from the advertising device 100.

As described above, the advertising device 100 optimizes the advertising content extraction process to extract the advertising content that is expected to have a high advertising effect for the user who has requested the advertisement distribution, and the extracted advertising content is extracted. This is an information processing device that delivers advertisement content.

Note that, as described above, the advertising device 100 identifies the user terminal 10 when delivering the advertising content and identifies the user terminal 10 that delivers the advertising content. For example, the user can be identified by including the user identification information in a cookie exchanged between the web browser of the user terminal 10 and the advertising device 100. However, the method of identifying the user is not limited to the above. For example, a dedicated program may be set in the user terminal 10 and the user identification information may be transmitted to the advertising device 100 from the dedicated program.

[3. Configuration of advertising device]
Next, the configuration of the advertising device 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating a configuration example of the advertising device 100 according to the embodiment. As shown in FIG. 3, the advertising device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The advertising device 100 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator who uses the advertising device 100, and a display unit (for example, a liquid crystal display) for displaying various information. You may have.

(About communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. The communication unit 110 is connected to the network N by wire or wirelessly, and transmits and receives information to and from the user terminal 10, the advertiser terminal 20, and the web server 30 via the network N.

(About storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. The storage unit 120 includes an advertisement information storage unit 121, a user information storage unit 122, a distribution log storage unit 123, and a model storage unit 124.

(About advertisement information storage unit 121)
The advertisement information storage unit 121 stores information about advertisement content submitted from the advertiser terminal 20. Here, FIG. 4 illustrates an example of the advertisement information storage unit 121 according to the embodiment. FIG. 4 is a diagram illustrating an example of the advertisement information storage unit 121 according to the embodiment. In the example shown in FIG. 4, the advertisement information storage unit 121 has items such as “advertiser ID”, “advertisement group ID”, “advertisement content ID”, “designated information”, and “delivery result”. Further, the item of “designated information” further includes items such as “lower limit number of distributions” and “designated total number of distributions”. Further, the item of “delivery result” further includes items such as “number of individual deliveries”, “individual CTR”, “total delivery number” and “total CTR”.

The “advertiser ID” indicates identification information for identifying the advertiser or the advertiser terminal 20. The “advertisement group ID” indicates identification information for identifying a group for each of a plurality of advertisement contents when an advertiser submits a plurality of advertisement contents. Advertising content belonging to the same advertising group has a common advertising target. The “advertising content ID” indicates identification information for identifying the advertising content submitted from the advertiser to the advertising device 100.

In addition, below, the identification information shown in FIG. 4 may be used as a reference code. For example, the advertiser identified by the advertiser ID "CL11" is identified by "advertiser CL11", the advertisement group identified by the advertisement group ID "AD11" is identified by "advertisement group AD11", and the advertisement content ID "C111" is identified. The advertising content to be displayed may be referred to as “advertising content C111”.

The “designated information” indicates information designated by the advertiser at the time of submission. The “lower limit number of distributions” indicates the lower limit number of distributions set for each advertisement content. That is, when the advertising content is submitted by the advertiser by the impression number guarantee type (that is, the submission method that specifies the number of delivery of the advertising content to the user), the advertising device 100 distributes the lower limit for each advertising content. Accepts a number specification. The “designated total delivery number” indicates the guaranteed delivery number in the entire advertisement group. That is, the advertising device 100 performs the distribution process so that the total number of distributions of each advertisement content reaches the designated total number of distributions.

The “delivery result” indicates information acquired after the advertisement content is delivered. The “individual delivery number” indicates the number of each advertisement content actually delivered to the user. The “individual CTR” indicates a numerical value of the CTR calculated based on the number of individual distributions and the number of times each advertising content is clicked by the user. The CTR is calculated by "click count/impression count (display count of advertisement content)".

The “total number of distributions” indicates the total number of distributions of each advertisement content belonging to the same advertisement group. “Total CTR” indicates a numerical value of CTR calculated based on the total number of distributions and the number of times each advertising content belonging to the same advertising group is clicked. Note that the distribution information item is blank for advertisement content that has no distribution record at this time.

That is, in the example illustrated in FIG. 4, the advertiser identified by the advertiser ID “CL11” submits a plurality of advertisement contents as an advertisement group identified by the advertisement group ID “AD11”. Further, the advertisement group AD11 includes the advertisement contents C111 to C114, and for example, the lower limit distribution number set in the advertisement content C111 is “100000” times. The total number of distributions in the advertisement group AD11 designated by the advertiser CL11 is “1000000”. Also, each advertisement content belonging to the advertisement group AD11 has already been distributed. For example, the number of individual distributions of the advertisement content C111 is “51000” times, and the individual CTR is “0.012”. The total number of distributions in the entire advertisement group AD11 is “183000”, and the total CTR is “0.01”.

It should be noted that the data of the advertisement content (video content or still image content) actually distributed to the user terminal 10 may be stored in a predetermined advertisement distribution server provided separately from the advertising device 100. In this case, the advertising device 100 identifies the advertising content stored in the external advertising distribution server based on the advertising content ID stored in the advertising information storage unit 121. Then, the advertising device 100 controls the advertisement distribution server to distribute the specified advertising content to the user terminal 10.

Further, the advertising device 100 may accept a predetermined ratio as the lower limit distribution number, instead of specifying the number itself. Specifically, the advertising device 100 can receive the designation that each advertising content is distributed for at least 10% of the designated total number of distributions. Further, the advertising device 100 may accept only the specified total number of deliveries and may not accept the individual lower limit number of deliveries.

(Regarding the user information storage unit 122)
The user information storage unit 122 stores information about the user who is the distribution target of the advertising content. Here, FIG. 5 illustrates an example of the user information storage unit 122 according to the embodiment. FIG. 5 is a diagram illustrating an example of the user information storage unit 122 according to the embodiment. In the example shown in FIG. 5, the user information storage unit 122 has items such as “user ID” and “user information”. Further, the “user information” includes items such as “name”, “sex”, “age”, and “place of residence”.

The “user ID” is identification information that identifies the user terminal 10 and the user. Note that the user ID is assumed to match the reference numeral of the user who operates the user terminal 10. That is, it is assumed that the user terminal 10 identified by the user ID “U11” is a terminal device operated by the user “U11”.

"Name" indicates the name of the user. “Gender” indicates the gender of the user. “Age” indicates the age of the user. Note that, as shown in FIG. 5, no specific numerical value is shown in the "age" item, and an age group such as "20s" may be shown. The "residential place" indicates the place of residence of the user.

For example, in FIG. 5, the name of the user U11 identified by the user ID “U11” is “AAA”, the gender is “male”, the age is “20s”, and the place of residence is “A prefecture”. ] Is an example.

Although not shown in FIG. 5, the user information storage unit 122 may store user behavior information such as a website access history and a shopping site purchase history. In addition, the user information storage unit 122 may store a device ID, which is information for identifying the user terminal 10 itself. Unlike the user ID, the device ID indicates identification information recorded in the terminal device itself. Further, the user information storage unit 122 may include items such as “annual income” indicating the user's annual income and “preference” indicating the user's preference, as the user's attribute information.

(About delivery log storage unit 123)
The distribution log storage unit 123 stores a log regarding distribution of advertisement content. Here, FIG. 6 illustrates an example of the delivery log storage unit 123 according to the embodiment. FIG. 6 is a diagram illustrating an example of the delivery log storage unit 123 according to the embodiment. In the example illustrated in FIG. 6, the delivery log storage unit 123 is configured from a group of log information including items such as “delivered advertisement content ID”, “delivery destination user ID”, and “date and time when advertisement content is delivered”. The configured user information file F01 is stored. For example, the log forming the user information file F01 is added by being written in the user information file F01 at the timing when the advertising content is distributed.

The “advertising content ID” indicates identification information that identifies the advertising content. In FIG. 6, the advertisement content ID is described as “AD=C111”. This indicates that the advertising content delivered to the user is the advertising content identified by “C111”. The advertising content ID corresponds to the advertising content ID shown in FIG.

The “user ID” indicates identification information that identifies the user terminal 10 to which the advertising content has been distributed. In FIG. 6, the user ID is described as “UserID=U11”. This indicates that the user to whom the advertising content is distributed is the user identified by “U11”. The user ID corresponds to the user ID shown in FIG.

The "date and time when the advertisement content is distributed" is represented by a numerical value such as "2015/04/10_20:30" in FIG.

For example, in an example of the log forming the user information file F01 shown in FIG. 6, the advertisement content C111 identified by the advertisement content ID “C111” is “April 10, 2015 20:30” and the user ID “ "U11" indicates that the content has been delivered to the user terminal 10.

It should be noted that the distribution log storage unit 123 may not be inside the advertising device 100 but may be, for example, a predetermined log storage server connected to the outside. In this case, the acquisition unit 133 described later acquires the log stored in the predetermined log storage server via the network N.

(About the model storage unit 124)
The model storage unit 124 stores a log regarding distribution of advertisement content. Here, FIG. 7 illustrates an example of the model storage unit 124 according to the embodiment. FIG. 7 is a diagram illustrating an example of the model storage unit 124 according to the embodiment. In the example shown in FIG. 7, the model storage unit 124 has items such as “advertisement group ID”, “advertisement content ID”, and “model ID”.

The “advertisement group ID” and the “advertisement content ID” correspond to the same item shown in FIG. The “model ID” indicates information that identifies a prediction model that predicts the advertising effect. As shown in FIG. 7, the prediction model is created for each advertising content.

For example, FIG. 7 illustrates an example in which the prediction model created for the advertising content “C111” included in the advertising group “AD11” is “M111”.

(Regarding the control unit 130)
In the control unit 130, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like allows various programs (corresponding to an example of an extraction program) stored in a storage device inside the advertising device 100 to use a RAM as a work area. It is realized by being executed as. Further, the control unit 130 is realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As illustrated in FIG. 3, the control unit 130 includes a submission reception unit 131, a reception unit 132, an acquisition unit 133, an extraction unit 134, a distribution unit 135, and a notification unit 136, which will be described below. It realizes or executes information processing functions and actions. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 3, and may be another configuration as long as it is a configuration for performing information processing described later. Further, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship illustrated in FIG. 3 and may be another connection relationship.

(About the submission reception part 131)
The submission acceptance unit 131 accepts submission of advertisement content from the advertiser terminal 20. Specifically, the submission reception unit 131 receives submission of a plurality of advertisement contents related to a common advertisement target. Then, the submission acceptance unit 131 stores information regarding a plurality of submitted advertisement contents in the advertisement information storage unit 121 in association with the advertiser ID that identifies the advertiser of the submission source and the advertisement group ID. To do.

Further, the submission reception unit 131 receives the total number of distributions designated for each advertisement group and the lower limit number of distributions designated for each advertisement content. As a result, the advertiser can guarantee the number of impressions of the submitted advertisement and can prevent a situation in which only a specific advertisement content is not delivered from the plurality of advertisement content.

(About receiving unit 132)
The receiving unit 132 receives a distribution request for advertising content. Specifically, the receiving unit 132 receives a request transmitted from the user terminal 10 that displays a web page, which is a request regarding distribution of advertisement content displayed in an advertisement frame included in the web page.

In addition, the receiving unit 132 receives a request for advertisement distribution transmitted from the user terminal 10 and also receives information for identifying the user terminal 10. For example, the receiving unit 132 receives a cookie transmitted from the user terminal 10 as information for identifying the user terminal 10. Then, the reception unit 132 sends the received information to the acquisition unit 133 described below.

(About the acquisition unit 133)
The acquisition unit 133 acquires various types of information. For example, the acquisition unit 133 acquires information about the user, information about the advertisement content, and information about the advertisement distribution. Specifically, the acquisition unit 133 acquires information regarding distribution of advertisement content from the distribution log storage unit 123. In addition, the acquisition unit 133 acquires, from the web server 30 and the user terminal 10, information regarding the advertising effect such as the number of times the user has clicked on the actually distributed advertising content. At this time, the acquisition unit 133 acquires an index value indicating the advertising effect of a plurality of advertisement contents for each predetermined user group classified based on the attributes of the users who are the distribution targets of the advertisement contents.

In addition, the acquisition unit 133 acquires an index value indicating the advertising effect of a plurality of advertisement contents for each predetermined user group that is classified based on the attributes of the users who are the distribution targets of the advertisement contents. Here, the acquisition process of the index value by the acquisition unit 133 will be described with reference to FIG.

FIG. 8 is a diagram showing an example of the data table 125 according to the embodiment. Although not shown in FIG. 3, the data table 125 is stored in, for example, the advertisement information storage unit 121. As shown in FIG. 8, the data table 125 has items such as “advertisement content ID”, “click”, and “user information”. Further, the “user information” includes items such as “sex”, “age”, and “place of residence”.

The “advertising content ID” corresponds to the same item shown in FIG. “Click” indicates whether or not the distributed advertising content has been clicked. When it is clicked, "1" is described in the item of "click", and when it is not clicked, "0" is described in the item of "click".

“User information” indicates information about the user to whom the advertising content is distributed. The items of “sex”, “age”, and “place of residence” correspond to the same items shown in FIG. In the data table 125, each item is represented by a numerical value. For example, in the item of “sex”, “male” is expressed as “0” and “female” is expressed as “1”. In the “age” item, “20s” is expressed as “1”, “40s” is expressed as “2”, and “60s” is expressed as “3”. In the item of “place of residence”, “A prefecture” is expressed as “1”, “B prefecture” is expressed as “2”, and “C prefecture” is expressed as “3”. The actual user information includes age groups other than the above in the item of age, and the above-mentioned number of residences in the item of residence, which will be omitted for simplification of description.

That is, in the data table 125, information about what attribute the advertising content has been distributed to, and clicked on is stored. The acquisition unit 133 acquires the CTR corresponding to each user group using this information. For example, the acquisition unit 133 acquires the CTR of the user group corresponding to the age “1” based on the number of distributions and the number of clicks distributed to the user group corresponding to the age “1”. The acquisition unit 133 can acquire the CTR for a user group that is arbitrarily set based on the user attribute. For example, the acquisition unit 133 extracts the data whose age is “1” and whose gender is “0”, and acquires the CTR of the user group classified by the age “1” and the gender “0”. To do. The index value acquired by the acquisition unit 133 is used in the extraction process of the extraction unit 134 described below.

The acquisition unit 133 also stores the acquired information in a predetermined storage unit. For example, the acquisition unit 133 stores information regarding distribution of advertising content and information regarding advertising effectiveness of each advertising content in the advertising information storage unit 121. In addition, the acquisition unit 133 stores the acquired user information in the user information storage unit 122.

(About the extraction unit 134)
The extraction unit 134 extracts the advertising content to be delivered to the user from the plurality of advertising content based on the correlation between the index value acquired by the acquisition unit 133 and the user group.

After the advertisement content is submitted, before the advertisement content is actually delivered, or when the index value indicating the advertisement effect due to the delivery is not sufficiently acquired, the extraction unit 134 determines the advertisement group. The advertisement content is randomly extracted until a predetermined ratio of the total number of deliveries designated in (1) is reached.

Then, when the acquisition unit 133 acquires the distribution information of the advertising content and the information regarding the index value indicating the advertising effect of the advertising content is accumulated, the extraction unit 134 extracts the advertising content using the information. .. Hereinafter, two different types of optimization of the extraction processing executed by the extraction unit 134 will be described.

As a first method, the extraction unit 134 performs extraction optimized as follows. As described above, the acquisition unit 133 acquires the index value indicating the advertising effect of each advertising content for each predetermined user group classified based on the attribute of the user. Therefore, the extraction unit 134 determines which user group the user who has transmitted the advertisement distribution request belongs to. Then, the extraction unit 134 preferentially extracts the advertisement content having the highest CTR of the user group corresponding to the user among the plurality of advertisement contents in the same advertisement group.

Specifically, the extraction unit 134 classifies the user group into three elements, “age, sex, place of residence”. In this case, the extraction unit 134 classifies the users who have transmitted the advertisement distribution request by “age, sex, place of residence” and determines which user group they belong to. For example, the extraction unit 134 classifies the user U11 illustrated in FIG. 5 as belonging to the user group of (sex, age, place of residence)=(0, 1, 1). Similarly, the extraction unit 134 classifies the user U12 as belonging to the user group of (0, 2, 2).

Then, when extracting the advertising content for the user U11, the extracting unit 134 has the highest CTR among a plurality of advertising content belonging to the same advertising group among those actually distributed to the user group of (0, 1, 1). Prioritize the extraction of high advertising content. In this way, the extraction unit 134 refers to the correlation between the index value of the randomly distributed advertisement content and the user group, and as a result, extracts the advertisement content with the highest CTR. Accordingly, the extraction unit 134 can extract the advertising content that is assumed to have the highest advertising effect, according to the user who requested the advertisement distribution.

Next, the second method will be described. In the first method, the extraction unit 134 has shown an example in which the extraction process is optimized by comparing the CTRs of the classified user groups. In this case, it may happen that the advertising content cannot be properly extracted for the user group whose distribution record is insufficient. For example, when the user information is extremely fragmented and a large number of user groups are formed, when the predetermined number of random distributions are completed, the number of samples that the user groups match is small and a reliable CTR cannot be obtained. There are cases.

Therefore, the extraction unit 134 uses a prediction model created based on the correlation between the index value and the user group to extract the advertisement content to be delivered to the user from the plurality of advertisement content. The extraction unit 134 creates a prediction model for each advertisement content.

For example, the extraction unit 134 is assumed to classify the user group into three elements, “age, sex, residence”. At this time, the extraction unit 134 further classifies the user group for each one type of attribute, which is an element for classifying the user group, such as "age", "sex", and "residential place" (for convenience of description, "small user"). Group"). Then, the extraction unit 134 creates, for example, the following formula (1) in order to calculate the weight of the index value for each small user group in the entire user group.

y _(c111) = ω ₁ ·x ₁ +ω ₂ ·x ₂ +ω ₃ ·x ₃ ... (1)

The above formula (1) is a formula showing a model for predicting the CTR for each advertisement content for the user. In the above formula (1), a model corresponding to the advertising content C111 is shown as an example of the prediction model. Such a prediction model corresponds to the model M111 shown in FIG.

In the above formula (1), “y” is a predicted CTR calculated when “x ₁ , x ₂ , x ₃ ” is substituted with a numerical value corresponding to “age, sex, residence” of a predetermined user. Show. “X” indicates a CTR corresponding to a small user group, which is a user group classified by element. Specifically, “x ₁ ”corresponds to “age”, “x ₂ ”corresponds to “sex”, and “x ₃ ”corresponds to “place of residence”. “Ω” is a coefficient of “x” and indicates a predetermined weight value. Specifically, “ω ₁ ”is a weight value of “x ₁ ”, “ω ₂ ”is a weight value of “x ₂ ”, and “ω ₃ ”is a weight value of “x ₃ ”. Is. As described above, the prediction model includes a variable (for example, “ω ₁ ·x ₁ ”) including the index value “x” corresponding to the user group classified based on one type of attribute and the predetermined weight value “ω”. _] )) are combined. For example, when the attribute of the user U11 shown in FIG. 5 is applied to the above expression (1), the above expression (1) can be described as the following expression (2).

y _{(U11, c111)} = ω ₁ ·(CTR of the advertising content C111 at the age of the user U11) + ω ₂ ·(CTR of the advertising content C111 according to the gender of the user U11) + ω ₃ ·(the residence of the user U11) CTR of advertisement content C111 in...) (2)

More specifically, the attribute “20s, male, A prefecture” of the user U11 is substituted into the above equation (2). That is, the CTR predicted for the user U111 is “CTR in the advertising content C111 of the user in his 20s”, “CTR in the advertising content C111 of the male user”, and “CTR in the advertising content C111 of the A prefecture user”. , Each CTR is derived by a value calculated by multiplying a predetermined weight value.

The value corresponding to the predetermined weight value “ω” is optimized as the regression analysis progresses. The calculation of the weight value “ω” will be described. As described above, the acquisition unit 133 can acquire information such as whether or not the advertising content C111 is clicked, and the CTR corresponding to each attribute. When the advertisement content C111 is clicked by the user in one trial of the advertisement distribution, the CTR becomes “1”. Similarly, when the advertisement content C111 is not clicked by the user in one trial, the CTR becomes “0”. Therefore, for example, the example of the data shown in FIG. 8 is expressed by the following equations (3) and (4).

1 = ω ₁ · (CTR of 20's user group) + ω ₂ · (CTR of male user group) + ω ₃ · (CTR of A prefecture user group) (3)
0 = ω ₁ · (CTR of 60's user group) + ω ₂ · (CTR of male user group) + ω ₃ · (CTR of B prefecture user group) (4)

In the above, two pieces of data are shown as an example, but in reality, each time the advertisement distribution is tried, the same expressions as the above expressions (3) and (4) are generated. That is, the extraction unit 134 generates the equations corresponding to the equations (3) and (4) each time the information regarding the distributed advertisement content C111 is acquired by the acquisition unit 133, and executes the regression analysis. Then, the extraction unit 134 calculates “ω” that satisfies the above formulas (3) and (4) at the time when the random distribution process ends. Thereby, the extraction unit 134 can create a prediction model corresponding to the advertising content C111. In the calculation process using the above equations (3) and (4), the left side is not “1” or “0” itself, but a predetermined error is assumed, and the value obtained by squaring the difference from the error is the minimum. The optimum solution of “ω” may be calculated using a method such as a least squares method that approximates the value.

Then, the extraction unit 134 applies the attribute information of the user who has requested the advertisement distribution to the created prediction model for each advertisement content, thereby calculating the predicted CTR for the user. Then, the extraction unit 134 extracts the advertising content having the highest calculated predicted CTR within the same advertising group. According to such a method, even when the number of CTR samples that satisfy all the factors of “20s, men, A prefecture” is not sufficient, each small user group (“20s”, “male”, “A”) The CTR of “prefecture” can be used to obtain the predicted CTR for a user who satisfies “20s, males, A prefecture”.

As described above, in the optimization using the prediction model, the acquisition unit 133 acquires the index value corresponding to the user group classified based on one type of attribute of the user. Then, the extraction unit 134 uses a prediction model created by combining variables including an index value corresponding to a user group classified based on one type of attribute and a predetermined weight value, and By applying the attributes to the prediction model, the advertising content predicted to have the highest index value for the user is extracted.

As described above, the extraction unit 134 is optimized for the user group by using the correlation of each attribute information of the user even in the situation where the advertisement content is actually distributed to the user group in a small number. Advertising content can be extracted. Further, the extraction unit 134 can recursively calculate the weight value “ω” of the prediction model, so that the influence of each attribute in the prediction model can be set to an appropriate weight. For example, consider an example where the residence, which is one of the attributes, is classified into a large number instead of the small number as described above (for example, in the case of being classified into 47 prefectures). In this case, a sufficient sample may not be obtained in the process of randomly distributing the advertising content to the user. For example, if the advertisement content is distributed “twice” to “user in A prefecture”, and the advertisement content is clicked twice, the CTR of “user in A prefecture” is “1”. Due to such a small number of samples, if an extreme value is observed, it will affect the prediction model and reduce the reliability. However, by calculating the weight value recursively as described above, the advertising device 100 can suppress the influence of such an extreme value on the prediction model (in this case, the weight value “ω ₃ ”). , 0, or a negative value is assumed to avoid the influence on the entire prediction model).

Note that the extraction unit 134 may adjust the ratio at which the advertising content is extracted in a predetermined case even after the extraction process is optimized. For example, the extraction unit 134 may extract only a small number of advertisement contents from the plurality of advertisement contents when the optimized extraction process is executed for all the advertisement distribution requests. For example, when there is a significant difference in advertising effect among a plurality of advertising contents, the extraction unit 134 may extract only the advertising content having the highest advertising effect. Therefore, the delivery unit 135, which will be described later, may not be able to achieve the delivery number designated by the advertiser for advertisement content with a low CTR.

Therefore, the extraction unit 134 may perform processing such as randomly extracting a plurality of advertisement contents from the received request for advertisement distribution, for example, until the lower limit number of distributions designated by the advertiser is reached. As a result, the extraction unit 134 can prevent the number of distributions of the specific advertisement content from being extremely reduced.

(About distribution unit 135)
The distribution unit 135 distributes the advertising content extracted by the extraction unit 134 to the user terminal 10 corresponding to the user who has transmitted the distribution request for the advertising content received by the receiving unit 132.

Note that, as described above, the data itself of the advertising content that is actually distributed may not be stored in the advertising information storage unit 121 of the advertising device 100. For example, the delivery unit 135 may deliver the advertisement content extracted by the extraction unit 134 to the user terminal 10 by transmitting an advertisement delivery control command to a predetermined advertisement delivery server provided outside.

(About the notification unit 136)
The notification unit 136 notifies the advertiser of information regarding advertisement distribution. Specifically, the notification unit 136 notifies the advertiser of an index value indicating an advertising effect such as the number of distributions of the advertising content and the number of times the advertising content is clicked by the user.

In addition, the notification unit 136 may notify, based on the result of actual distribution, which user group has a high appeal effect in the elements extracted from the advertising content. Specifically, the notification unit 136 determines that the CTR corresponding to the group of users in their twenties, females, regarding the advertising content produced based on “blue” among the plurality of advertising content belonging to the same advertising group. Notify that it is higher than the CTR of other advertising content. In other words, the notification unit 136 notifies the advertiser of a reaction (for example, an advertising effect) from the user group due to the difference between the color that is an element that constitutes the content, the expression of the text in the content, and the characters. As a result, the advertiser can know the advertising content having a high appealing effect on the target user group and the advertising content preferred by the user. Note that, regarding the elements of the advertising content as described above, for example, when the advertising content is moving image content, the notification unit 136 uses known analysis methods such as analysis of pixels forming a moving image and analysis of text. By this, the elements that compose the content are automatically extracted. In addition, the notification unit 136 may analyze and extract the elements that constitute the advertising content based on the information (characters, shooting location, etc.) tagged for the video content.

[4. Extraction processing procedure]
Next, a procedure of extraction processing by the advertising device 100 according to the embodiment will be described with reference to FIGS. 9 and 10. First, with reference to FIG. 9, a flow of processing of randomly extracting advertisement content from a plurality of advertisement content will be described. FIG. 9 is a flowchart showing the extraction processing procedure (1) by the advertising device 100 according to the embodiment.

As shown in FIG. 9, the submission acceptance unit 131 accepts submission of advertisement content (step S101). Then, the receiving unit 132 determines whether or not a request for advertisement distribution has been received from the user terminal 10 (step S102). When the reception unit 132 has not received the advertisement distribution request (step S102; No), the reception unit 132 waits until the reception.

On the other hand, when the advertisement distribution request is received (step S102; Yes), the reception unit 132 sends information for identifying the user terminal 10 that has transmitted the advertisement distribution request to the acquisition unit 133. Then, the acquisition unit 133 acquires user information regarding the user terminal 10 that has transmitted the request regarding advertisement distribution (step S103). Then, the acquisition unit 133 sends the acquired user information and information regarding the advertisement distribution request to the extraction unit 134.

The extraction unit 134 randomly extracts the advertising content from the plurality of submitted advertising content (step S104). Then, the distribution unit 135 distributes the extracted advertising content to the user (step S105).

Then, the extraction unit 134 determines whether or not a predetermined number of advertisement contents have been distributed (step S106). When the predetermined number of advertisement contents are not distributed (step S106; No), the extraction process is not optimized, and the process proceeds to step S102. On the other hand, when the predetermined number of advertisement contents are distributed (step S106; Yes), the extraction unit 134 optimizes the extraction process (step S107).

Although the process of determining whether or not a predetermined number of advertisement contents has been distributed in step S106 has been described, step S106 determines whether or not a predetermined number of advertisement effect measurement results are acquired, not the distribution number. You may judge. That is, in step S106, whether or not to optimize the extraction process may be determined based on the fact that the information used for the optimization process has been acquired.

Next, the extraction process after the optimization has been performed will be described. FIG. 10 is a flowchart showing the extraction processing procedure (2) by the advertising device 100 according to the embodiment.

Since step S201 and step S202 correspond to step S102 and step S103 shown in FIG. 9, description thereof will be omitted. Then, the extraction unit 134 extracts the advertising content corresponding to the acquired user information (step S203). Here, the extraction unit 134 uses the optimized extraction process to extract the advertising content that is expected to have the highest advertising effect for the user.

Then, the distribution unit 135 distributes the extracted advertising content to the user (step S204). Then, the delivery unit 135 determines whether or not the total number of advertisement contents delivered to the user has reached the designated total delivery number designated by the advertiser (step S205). When the designated total number of deliveries has not been reached (step S205; No), the process transitions to step S201 to receive an advertisement delivery request from the user.

On the other hand, when the number of deliveries reaches the designated total number of deliveries (step S205; Yes), the notification unit 136 notifies the advertiser of the fact that the advertisement delivery has ended, the result of the advertisement delivery, etc. (step S206). ).

[5. Modification example)
The advertising device 100 described above may be implemented in various different forms other than the above-described embodiment. Therefore, other embodiments of the advertising device 100 will be described below.

[5-1. Index value)
In the above-described embodiment, the acquisition unit 133 has shown an example in which the CTR is acquired as the index value indicating the advertising effect of the advertising content. However, the acquisition unit 133 may acquire an index value indicating the advertising effect, in addition to the CTR.

For example, the acquisition unit 133 may acquire a CVR (Conversion Rate) indicating a rate at which the advertiser has obtained some profit (conversion) from the distribution of the advertising content. Further, when the advertising content is moving image content, the acquisition unit 133 may acquire the completion rate indicating the rate at which the advertising content is viewed to the end. In addition, the acquisition unit 133 may acquire the staying time during which the line of sight is retained in the advertisement content by an eye tracking method that analyzes the movement of the line of sight on the user's website. In this way, the acquisition unit 133 can acquire various information as the index value indicating the advertising effect.

Then, the extraction unit 134 extracts the advertising content delivered to the user based on the index values indicating the various advertising effects acquired by the acquisition unit 133. For example, the extraction unit 134 extracts, based on the index value for each user group, the advertising content that is expected to have a high appealing effect on each user group.

In this way, the advertising device 100 can acquire various index values indicating advertising effectiveness, not limited to CTR. As a result, the advertising device 100 can analyze the users who are distribution candidates of the advertising content in more detail, and thus it becomes easier to extract the advertising content having a high appeal effect, and as a result, the advertising effect of the advertising content to be distributed can be improved. Can be improved.

[5-2. Acquisition of data]
In the extraction process according to the above-described embodiment, the acquisition unit 133 has shown an example in which the user information is acquired when the user terminal 10 accesses the website provided by the web server 30. However, the acquisition unit 133 may acquire user information from the user terminal 10 without being limited to the case where the user terminal 10 accesses the website provided by the web server 30. This point will be described below.

For example, when the website accessed by the user terminal 10 is provided from a predetermined web server managed by the same management device as the management device that manages the advertising device 100 (for example, a front end server for the advertising device 100), The acquisition unit 133 can acquire user information from the user terminal 10. That is, the user terminal 10 transmits the user information of the user terminal 10 to the management device when accessing a website provided by a predetermined web server managed by the management device. This is realized by the user terminal 10 transmitting a cookie to the management device. In this case, the user information transmitted by the user terminal 10 is transmitted to the advertising device 100 via the management device. Accordingly, the acquisition unit 133 according to the advertising device 100 can acquire the user information of the user terminal 10 without going through the web server 30. Regarding the acquired user information, the acquisition unit 133 relates to the user information by collating a cookie transmitted from the user terminal 10 described above, or a device ID preset in the user terminal 10. Each user terminal 10 can be identified.

[5-3. Relationship with web page]
In the above-described embodiment, the submission acceptance unit 131 accepts submission of a plurality of advertisement contents related to a common advertising target, and the extraction unit 134 selects from among the plurality of advertisement contents accepted by the submission acceptance unit 131. An example of extracting the advertising content delivered to the user has been shown. Here, the extraction unit 134 may extract the advertising content based on a predetermined condition received from the advertiser.

For example, some advertisers may specify the content (for example, a web page, an application, etc.) for displaying the advertising content submitted by the advertiser. Specifically, the advertiser can specify conditions so that the web page provided by the competitor does not display the advertising content submitted by the advertiser. Alternatively, the advertiser may want to display his/her advertising content on the content in which information of a specific category is posted.

In this case, the extraction unit 134 may extract the advertising content in consideration of the conditions designated by the advertiser in addition to the processing optimized as described above. For example, an advertiser may publish advertisement content desired to be posted on a "news site", advertisement content desired to be posted on a "sports site", or a "shopping site" with respect to a plurality of advertisement content having a common advertisement target. The conditions relating to the web page on which the advertising content is distributed and displayed, such as the desired advertising content, are designated. Then, the extraction unit 134 extracts the advertisement content corresponding to various websites from the plurality of advertisement content having the common advertisement target according to the condition designated by the advertiser. Then, the distribution unit 135 distributes the advertising content extracted by the extraction unit 134 to the user.

That is, the advertising device 100 may accept designation of conditions regarding a web page or the like to which the advertising content is distributed from the advertiser in advance. Even in this case, the advertising device 100 can optimize the advertising effect of the delivered advertising content based on the information about the delivered advertising content. As a result, the advertising device 100 can deliver the highly appealing advertising content to more users while satisfying the advertiser's request.

[5-4. User group)
In the embodiment described above, the advertising device 100 has shown an example in which the users are classified into a predetermined user group based on the attribute information of the users. For example, in the embodiment, the advertising device 100 shows an example in which the users are classified into user groups based on “age”, “sex”, “residential place”, and the like. Here, the advertising device 100 may classify users by using different user information. For example, the advertising device 100 may classify users by using “annual income”, “preference”, or the like in the user information. For example, the advertising device 100 can perform a process of classifying users into a user group of “annual income of 10 million yen or more” and extracting advertisement content having a high advertising effect for users belonging to the user group. Further, as described above, the advertising device 100 can arbitrarily set the user group by using each element of the user information. As a result, the advertising device 100 can deliver advertising content with a high advertising effect to each of a wide range of users.

[5-5. Prediction model)
In the embodiment described above, first, the advertising device 100 randomly distributes a predetermined number of advertisement contents, and optimizes the extraction process when the predetermined number of distributions are completed. Here, the advertising device 100 may update the optimization process by continuing the random distribution for a predetermined ratio even when the random distribution ends.

That is, the extraction unit 134 randomly extracts the advertising content distributed to the user for a predetermined ratio of the advertising content to be extracted. Then, the acquisition unit 133 further acquires the index value for each user group when the advertisement content randomly extracted by the extraction unit 134 is distributed to the user. Then, the extraction unit 134 uses the prediction model updated based on the index value acquired by the acquisition unit 133 to extract the advertising content delivered to the user from the plurality of advertising content.

For example, the extraction unit 134 creates models M111 to M114 for the advertisement group AD11 shown in FIG. 7 after distributing a predetermined number of advertisement contents C111 to C114. After that, the extraction unit 134 extracts any one of the advertisement contents C111 to C114 by fitting the attribute information of the user who has transmitted the advertisement distribution request into the models M111 to M114. At this time, the extraction unit 134 randomly extracts the advertising contents C111 to C114 for a predetermined ratio (for example, 10%) without using the models M111 to M114. Then, the acquisition unit 133 acquires the advertising effect of the distributed advertising contents C111 to C114. Then, the extraction unit 134 appropriately updates the models M111 to M114 based on the acquired advertising effect.

In this way, the advertising device 100 can prevent a situation in which, for example, the extracted advertising content is heavily weighted on a specific advertising content, by adding a predetermined random element to the extraction process. In addition, in the case where the type of the advertising content that is clicked changes depending on the time, the advertising device 100 mixes the random distribution of a predetermined ratio as described above, and thus the distribution of the advertising content in a timely manner is appropriately performed. You can do it.

[5-6. Delivery mode]
In the above-described embodiment, the advertising device 100 has shown an example in which the specified total number of distributions received is distributed for a plurality of advertising contents. Specifically, as in the example of FIG. 4, when the advertising device 100 receives the designated total number of deliveries “1000000” from the advertiser CL11, the total number of deliveries of the advertisement group AD11 achieves “1,000,000”. Deliver advertising content to. That is, the advertising device 100 shows an example in which the number of impressions (the number of times the user terminal 10 is displayed) is designated to deliver the advertising content (impression-guaranteed advertisement). However, the form of advertisement distribution performed by the advertising device 100 is not limited to the above.

That is, the advertising device 100 can accept various forms of advertisement distribution regarding distribution of a plurality of advertisement contents related to a common advertising target. For example, the advertising device 100 may accept designation of “the number of distributions delivered to a specific target” as the advertising content.

As an example, the advertising device 100 receives the number of deliveries delivered to UUs (unique users) as a specific target. In this case, the advertising device 100 does not simply count the number of times the advertising content is distributed, but counts the number of distributions “1” when the advertising content is distributed to one user. In other words, the advertising terminal 100 may not count the number of distributions when the same advertising content is distributed to the same user. In this way, the advertising device 100 can accept submission of a plurality of advertising contents in a delivery form that achieves the specified number of UUs. Accordingly, the advertising device 100 can prevent, for example, a situation where a specific user frequently browses a web page or the like and generates a large number of distribution opportunities, so that the opportunity for the advertising content to be distributed to a wide range of users is lost. .. In addition, the number of counting a specific object such as UU may be expressed as a reach number. That is, the advertising device 100 can perform not only the distribution corresponding to the number of impressions in a so-called impression-guaranteed advertisement but also the advertisement distribution in a form that guarantees the reach number.

Further, the advertising device 100 can receive the number of times of distribution to a browser installed in a predetermined user terminal 10 as a specific target. In this case, the advertising device 100 counts the number of distribution “1” when the advertising content is distributed to the browser installed in the predetermined user terminal 10. In other words, the advertising device 100 may not count the number of distributions when the same advertising content is newly distributed to the same browser to which the advertising content has already been distributed. For example, the advertising device 100 identifies the browser to which the advertising content is delivered, based on the identification information (cookie or the like) transmitted/received to/from the browser.

Further, the advertising device 100 can also deliver the advertising content of the contract in which the charge amount is calculated according to the displayed time, for example. That is, the advertising device 100 performs advertisement distribution in a form that achieves the display time designated by the advertiser. Specifically, the advertising device 100 delivers the advertising content to the user terminal 10, and sends the advertising content to the user terminal 10 until the total display time of the advertising content on the user terminal 10 of the delivery destination reaches the designated display time. Is repeatedly delivered.

[6. Hardware configuration]
Further, the advertising device 100 according to the above-described embodiment is realized by, for example, a computer 1000 configured as shown in FIG. Hereinafter, the advertising device 100 will be described as an example. FIG. 11 is a hardware configuration diagram illustrating an example of a computer 1000 that realizes the function of the advertising device 100. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, an HDD 1400, a communication interface (I/F) 1500, an input/output interface (I/F) 1600, and a media interface (I/F) 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 to control each unit. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 starts up, a program dependent on the hardware of the computer 1000, and the like.

The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. The communication interface 1500 receives data from another device via the communication network 500 (corresponding to the network N shown in FIG. 2) and sends the data to the CPU 1100, and the data generated by the CPU 1100 is transmitted via the communication network 500 to another. Send to device.

The CPU 1100 controls output devices such as a display and a printer and input devices such as a keyboard and a mouse via the input/output interface 1600. The CPU 1100 acquires data from the input device via the input/output interface 1600. In addition, the CPU 1100 outputs the data generated via the input/output interface 1600 to the output device.

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the program or data to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) and a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

For example, when the computer 1000 functions as the advertising device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the function of the control unit 130 by executing the program loaded on the RAM 1200. Further, the HDD 1400 stores the data in the storage unit 120. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from another device via the communication network 500.

[7. Other]
Further, of the processes described in the above embodiment, all or part of the processes described as being automatically performed may be manually performed, or the processes described as manually performed may be performed. All or part of the process can be automatically performed by a known method. In addition, the processing procedures, specific names, information including various data and parameters shown in the above-mentioned documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

Further, each component of each device shown in the drawings is functionally conceptual and does not necessarily have to be physically configured as shown. That is, the specific form of distribution/integration of each device is not limited to that shown in the figure, and all or part of the device may be functionally or physically distributed/arranged in arbitrary units according to various loads and usage conditions. It can be integrated and configured. For example, the acquisition unit 133 and the extraction unit 134 illustrated in FIG. 3 may be integrated. In addition, for example, the information stored in the storage unit 120 may be stored in an external storage device via the network N.

Further, for example, in the above-described embodiment, an example is shown in which the advertising device 100 performs a reception process of receiving a submission of advertising content, an extraction process of extracting the advertising content to be distributed, and a distribution process of distributing the advertising content. It was However, the advertising device 100 described above may be separated into a reception device that performs reception processing, an extraction device that performs extraction processing, and a distribution device that performs distribution processing. In this case, the reception device has a submission reception unit 131. The extraction device includes an acquisition unit 133 and an extraction unit 134. The distribution device also includes a reception unit 132, a distribution unit 135, and a notification unit 136. In this case, the processing by the advertising device 100 described above is realized by the extraction processing system 1 including each of the reception device, the extraction device, and the distribution device.

Further, the above-described embodiments and modified examples can be appropriately combined within a range in which the processing content is not inconsistent.

[8. effect〕
As described above, the advertising device 100 according to the embodiment includes the submission reception unit 131 (an example of a reception unit), the acquisition unit 133, and the extraction unit 134. The submission acceptance unit 131 accepts submission of a plurality of advertisement contents related to a common advertisement target. The acquisition unit 133 acquires an index value indicating the advertising effect of a plurality of advertising contents for each user group classified based on the attribute of the user who is the distribution target of the advertising contents. The extraction unit 134 extracts the advertising content to be delivered to the user from the plurality of advertising content based on the correlation between the index value acquired by the acquisition unit 133 and the user group.

As described above, the advertising device 100 according to the embodiment accepts submission of a plurality of advertising contents having a common advertising target, for example, intended for users in the target layer. Then, the advertising device 100 optimizes the extraction of the advertising content based on the user information and the information about the advertising effect, and delivers the most suitable advertising content for each user group. As a result, the advertising device 100 can distribute the advertising content having a high appeal effect to more users.

Also, the extraction unit 134 predicts that the index value will be the highest for the user among the plurality of advertisement contents by using the prediction model created for each advertisement content based on the correlation between the index value and the user group. Extract the advertising content that will be played.

As described above, the advertising device 100 according to the embodiment can extract the advertising content by using the prediction model for each advertising content. As a result, the advertising device 100 can appropriately extract the highly effective advertising content from the plurality of advertising contents and cause the user to deliver it.

The acquisition unit 133 also acquires an index value corresponding to a user group classified based on one type of attribute of the user. Then, the extraction unit 134 uses a prediction model created by combining variables including an index value corresponding to a user group classified based on one type of attribute and a predetermined weight value, and By applying the attributes to the prediction model, the advertising content predicted to have the highest index value for the user is extracted.

As described above, the advertising device 100 according to the embodiment creates the advertising content prediction model in a format including a variable for each attribute information. Thereby, according to the advertising device 100, even if there is no sample in which the attribute information of the user completely matches, it is possible to calculate the influence on the overall advertising effect by applying the prediction model for each attribute, and thus the advertisement for the user. The effect can be predicted appropriately.

Further, the extraction unit 134 randomly extracts the advertising content distributed to the user for a predetermined ratio of the advertising content to be extracted. Then, the acquisition unit 133 further acquires the index value for each user group when the advertisement content randomly extracted by the extraction unit 134 is distributed to the user. Then, the extraction unit 134 uses the prediction model updated based on the index value acquired by the acquisition unit 133 to extract the advertising content delivered to the user from the plurality of advertising content.

As described above, the advertising device 100 according to the embodiment can randomly extract advertising content for a predetermined ratio (for example, 10% of the whole) even after the prediction model is created. Accordingly, the advertising device 100 can prevent the advertising content extracted by using the prediction model from being fixed.

In addition, the acquisition unit 133 acquires, as an attribute of the user, an index value corresponding to a user group classified based on at least one of the age, sex, and place of residence of the user.

In this way, the advertising device 100 according to the embodiment classifies users into user groups by using various types of user information. As a result, the advertising device 100 can classify the user group in detail, and thus can easily extract the advertising content that is expected to have a higher advertising effect.

In addition, the acquisition unit 133 uses, as the index value, the number of times the advertising content is selected by the user, information about the profit obtained by the advertiser of the advertising content by displaying the advertising content, and the number of times the advertising content is viewed by the user. , An index value calculated based on any one of the lengths of time the advertisement content is viewed by the user.

In this way, the advertising device 100 according to the embodiment can acquire the index value of the advertising effect indicated by various methods. Therefore, the advertising device 100 can optimize the extraction of the advertising content from various aspects.

As described above, some of the embodiments of the present application have been described in detail based on the drawings, but these are examples, and various modifications based on the knowledge of those skilled in the art, including the modes described in the section of the disclosure of the invention, It is possible to implement the present invention in other forms with improvements.

Further, the "section (module, unit)" described above can be read as "means" or "circuit". For example, the acquisition unit can be read as an acquisition unit or an acquisition circuit.

DESCRIPTION OF SYMBOLS 1 extraction processing system 10 user terminal 20 advertiser terminal 30 web server 100 advertising device 110 communication unit 120 storage unit 121 advertisement information storage unit 122 user information storage unit 123 delivery log storage unit 130 control unit 131 submission acceptance unit 132 reception unit 133 Acquisition unit 134 Extraction unit 135 Distribution unit 136 Notification unit

Claims (11)

A reception unit that receives the submission of multiple advertising contents related to a common advertising target,
Distribute any of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each of the plurality of user groups classified based on the attribute of the user who is the target of the distribution of the advertisement content. An acquisition unit that acquires an index value indicating the advertising effect obtained by
An advertisement delivered to a user belonging to any one of the plurality of user groups from the plurality of advertisement contents based on the correlation between the index value obtained by the obtaining unit and any one of the plurality of user groups An extraction unit for extracting content,
An extraction device comprising: A reception unit that receives the submission of a plurality of advertisement contents showing different characteristics of the common advertisement target,
An advertisement obtained by distributing one of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each user group classified based on the attribute of the user to whom the advertisement content is distributed. An acquisition unit that acquires an index value indicating the effect,
An extraction unit that extracts advertisement content to be delivered to the user from the plurality of advertisement contents based on the correlation between the index value acquired by the acquisition unit and the user group.
An extraction device comprising: The extraction unit is
Using the prediction model created for each of the advertising contents based on the correlation between the index value and the user group, the index value is predicted to be highest for the user among the plurality of advertising contents. Extract advertising content that
The extraction device according to claim 1 or 2, characterized in that. The acquisition unit is
Acquiring the index value corresponding to a user group classified based on one type of attribute of the user,
The extraction unit is
Predicting each attribute of the user by using the prediction model created by combining variables including an index value corresponding to a user group classified based on the one type of attribute and a predetermined weight value By fitting the model, the advertising content predicted to have the highest index value for the user is extracted.
The extraction device according to claim 3, wherein: The extraction unit is
Randomly extracting the advertising content delivered to the user for a predetermined ratio of the advertising content to be extracted,
The acquisition unit is
Acquiring an index value for each user group when the advertising content randomly extracted by the extraction unit is delivered to the user,
The extraction unit is
Using the prediction model updated based on the index value acquired by the acquisition unit, extract the advertising content to be delivered to the user from the plurality of advertising content.
The extraction device according to any one of claims 1 to 4, characterized in that. The acquisition unit is
As the attribute of the user, obtain the index value corresponding to a user group classified based on at least one of the age, sex, and place of residence of the user,
The extraction device according to any one of claims 1 to 5, characterized in that The acquisition unit is
As the index value, the number of times the advertising content is selected by the user, information about the profit obtained by the advertiser of the advertising content by displaying the advertising content, the number of times the advertising content is viewed by the user, the Obtaining an index value calculated based on any one of the lengths of time the advertising content is viewed by the user,
The extraction device according to any one of claims 1 to 6, which is characterized in that. A computer-implemented extraction method,
A reception process for receiving submissions of multiple advertising contents related to a common advertising target,
Distribute any of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each of the plurality of user groups classified based on the attribute of the user who is the target of the distribution of the advertisement content. An acquisition step of acquiring an index value indicating the advertising effect obtained by
An advertisement delivered to a user belonging to any one of the plurality of user groups from among the plurality of advertisement contents based on the correlation between the index value obtained in the obtaining step and any one of the plurality of user groups An extraction step of extracting the content,
An extraction method characterized by including. A reception procedure for receiving submissions of multiple advertising contents related to a common advertising target,
Distribute any of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each of the plurality of user groups classified based on the attribute of the user who is the target of the distribution of the advertisement content. The acquisition procedure to acquire the index value that shows the advertising effect obtained by
An advertisement delivered to a user belonging to any one of the plurality of user groups from among the plurality of advertisement contents based on the correlation between the index value acquired by the acquisition procedure and any one of the plurality of user groups An extraction procedure to extract the content,
An extraction program characterized by causing a computer to execute. A computer-implemented extraction method,
A receiving step for receiving the submission of a plurality of advertisement contents showing different features of a common advertisement target,
An advertisement obtained by distributing one of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each user group classified based on the attribute of the user to whom the advertisement content is distributed. An acquisition process for acquiring the index value indicating the effect,
An extraction step of extracting advertisement content to be delivered to the user from the plurality of advertisement content based on the correlation between the index value acquired in the acquisition step and the user group;
An extraction method characterized by including. A procedure for accepting the submission of multiple advertising contents that show different characteristics of a common advertising target,
An advertisement obtained by distributing one of the plurality of advertisement contents to a predetermined ratio or more of the total number of distributions of the advertisement content for each user group classified based on the attribute of the user to whom the advertisement content is distributed. Acquisition procedure to acquire the index value showing the effect,
An extraction procedure for extracting advertisement content to be delivered to the user from the plurality of advertisement content based on the correlation between the index value acquired by the acquisition procedure and the user group;
An extraction program characterized by causing a computer to execute.

Images