JP6948298B2 - Information processing equipment, information processing methods, and programs - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program.

Traditionally, the data processor is the step of accessing the publisher log, which contains data describing past impressions provided for page views on the publisher site, where the data is the user identifier, publisher, for each impression. A step that includes a publisher identifier that identifies the page view, a page view value that identifies the page view, and multiple publisher key values for the publisher provided for the impression, and a data processor for each publisher from the publisher log. In the step of generating publisher pageview data, each publisher's publisher pageview data comprises a pageview record representing the pageviews that occurred during the first period, and each pageview record Represents one or more pageviews associated with a pair of user and issuer identifiers during the second period that occurred during the first period, and each pageview record is a user identifier, issuer identifier, And a step that contains the issuer key value provided for the impression generated for one or more pageviews represented by the pageview record, and a step that receives the issuer's segment definition in the data processor, a segment. A user identifier from a record that contains a step whose definition defines a segment by one or more issuer key values of the issuer, and an issuer key value that satisfies the segment defined by the segment definition by the data processor. To a segment, and to determine one or more segment characteristics based on the page view record of the user identifier assigned to the segment and the user identifier assigned to the segment by the data processor, and the segment definition. A computer practice method is known that comprises a step of providing response data describing one or more characteristics of a segment in response to (see Patent Document 1).

Japanese Patent Application Laid-Open No. 2013-503392

In the conventional technology, it may not be possible to flexibly define a group of users who share the same content settings.

The present invention has been made in consideration of such circumstances, and provides an information processing device, an information processing method, and a program capable of flexibly defining a group of users who share common settings related to contents. That is one of the purposes.

One aspect of the present invention is to classify the coding processing unit that performs coding processing on the identification information that identifies the user who browses the content and the result of performing the coding processing, and group the users. A grouping processing unit that outputs information of a group of users who share settings related to the content based on the grouping result is provided, and the coding processing unit is the same by changing the parameters of the coding processing. The identification information group of the above is encoded by a plurality of layers, and the grouping processing unit groups the users for each layer or groups the users based on the logical product of the classification results between the layers. It is an information processing device.

According to one aspect of the present invention, it is possible to flexibly define a group of users who share settings related to contents.

It is a figure which shows an example of the structure and use environment of an information processing apparatus 100. It is a figure which shows an example of the screen (advertisement placement surface) provided as the main content. It is a figure which shows an example of the content of the user information 38. It is a sequence diagram which shows an example of the process flow of a bucket test. It is a figure which shows typically the content of the processing by a coding processing unit 130. It is a figure which shows typically the processing content of the grouping processing unit 140. It is a flowchart which shows an example of the flow of processing executed in the information processing apparatus 100. It is a figure which shows typically an example of the process executed in the content providing apparatus 30.

Hereinafter, the information processing apparatus, the information processing method, and the embodiment of the program of the present invention will be described with reference to the drawings.

The information processing device is realized by one or more processors. The information processing device may be realized by one or more decentralized processors. The information processing device is a device that receives a request from a request source, groups users who browse the contents, and outputs information of a group of users who share settings related to the contents. The content-related settings are, for example, advertisement distribution settings embedded in the content. Hereinafter, this will be described as a premise. The request source and the output destination of the information may be a content providing device that provides the content to the user's terminal device, or an advertisement distribution device that delivers the advertisement. The information processing device may be a function of a content providing device or an advertisement distribution device, that is, a virtual device.

As a specific process, the information processing device performs a coding process on the identification information that identifies the user who browses the content, classifies the result of the coding process, and groups the users. The coding process is, for example, a process of inputting identification information into a hash function. Hereinafter, this will be described as a premise. The information processing apparatus performs coding processing on the same identification information group in a plurality of layers by changing the parameters (for example, seeds) of the hash function. Then, the information processing device groups the users for each layer, or groups the users based on the logical product of the classification results between the layers. As a result, the information processing device can flexibly define a group of users who share advertisement distribution settings (settings related to contents).

The identification information that identifies the user is, for example, a cookie for identifying a browser. Strictly speaking, a cookie is identification information of a terminal device, but in the present specification, it is treated as a kind of identification information of a user in a broad sense. The browser may include not only a browser that is started independently but also a browser (in-app browser) built in the application program. Further, the identification information is not limited to the cookie, and may be information such as a so-called user ID.

[Overview]
FIG. 1 is a diagram showing an example of the configuration and usage environment of the information processing device 100. In the illustrated example, one or more terminal devices 10, a content providing device 30, an advertisement distribution device 50, and an information providing device 100 are connected to a network NW. The network NW includes, for example, the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a radio base station, a provider terminal, a dedicated line, and the like.

The terminal device 10 is, for example, a mobile phone such as a personal computer or a smartphone, a tablet terminal, or the like. In these, UAs (User Agents) such as browsers and application programs are started. The UA of the terminal device 10 displays the content screen provided by the content providing device 30, and also transmits a request corresponding to the input operation made by the user of the terminal device 10 to the content providing device 30.

The content providing device 30 includes, for example, a web management unit 32 and an advertisement determination unit 34. Further, the content providing device 30 stores information such as the main content 36 and the user information 38 in a storage device such as a RAM (Random Access Memory), an HDD (Hard Disk Drive), or a flash memory. A combination of one or both of the web management unit 32 and the advertisement decision unit 34 is an example of the "distribution unit".

The web management unit 32 reads one main content from the main content 36 and transmits it to the terminal device 10 in response to a request from the UA of the terminal device 10. FIG. 2 is a diagram showing an example of a screen (advertisement placement surface) provided as the main content. An advertising space AD is provided on this screen together with news article frames NS (1) to NS (3). The advertisements displayed in the ad space AD include general-purpose advertisements for everyone and targeting advertisements targeting users who are viewing the main content. The number of ad space ADs in the main content may be constant or may differ depending on the distribution settings. The web management unit 32 attaches the advertisement to the web page and transmits it to the terminal device 10, or causes the advertisement to be transmitted to the terminal device 10 by redirect.

The advertisement determination unit 34 determines, for example, the ratio between the general-purpose advertisement and the targeting advertisement. In addition, the advertisement determination unit 34 refers to the user information 38 and determines the targeting advertisement. FIG. 3 is a diagram showing an example of the contents of the user information 38. The user information 38 includes a user ID, age, gender, occupation, address, office, interests, search history (query history), site browsing information, model used, etc., in response to the above-mentioned identification information cookie. Information is associated with it.

The user ID in the user information 38 is used when the user logs in to the service that provides the main content. Information such as the user's age, gender, occupation, place of employment, and matters of interest may be entered by the user at the time of registration, or may be estimated from the user's search history or site browsing history. You may. The search history and the site browsing history are extracted from the log information collected by the content providing device 30. The model to be used is acquired, for example, in the process of communicating with the terminal device 10.

The ad decision-making unit 34 uses site retargeting based on site browsing history, search targeting based on search history, targeting based on gender, age, region, etc., device targeting based on the model used, interest match based on what is of interest, and advertisement distribution. Targeting advertising is determined by a method such as placement targeting based on the information that specifies whether or not the targeting is possible.

As described above, there are various factors in the distribution setting in the advertisement distribution, such as the number of ad space ADs in the main content, the ratio between the general-purpose advertisement and the targeting advertisement, and the switching (or mixed ratio) of the targeting method. Therefore, the number of combinations of distribution settings is relatively large. The content providing device 30 has different distribution settings for each group of users (hereinafter, referred to as a segment) provided by the information processing device 100. The function of determining the distribution setting and the targeting advertisement (function as the "distribution unit") may be the function of the advertisement distribution device 50 instead of the content providing device 30.

The advertisement distribution device 50 extracts one or more contents from the advertisement contents 52 held in the storage device in response to the request from the content providing device 30, and transmits the contents to the content providing device 30 or the terminal device 10.

[Information processing device]
The information processing device 100 includes, for example, a communication unit 110, a seed reception unit 120, a coding processing unit 130, and a grouping processing unit 140. The components other than the communication unit 110 are realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program (software). Some or all of these components are hardware such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit) It may be realized by (including circuits), or it may be realized by the cooperation of software and hardware. The program may be stored in advance in a storage device (a storage device including a non-transient storage medium) such as an HDD (Hard Disk Drive) or a flash memory, or a removable storage device such as a DVD or a CD-ROM. It is stored in a medium (non-transient storage medium) and may be installed by mounting the storage medium in a drive device.

The communication unit 110 is, for example, a communication interface such as a NIC (Network Interface Card) for connecting to a network NW.

The seed receiving unit 120, the coding processing unit 130, and the grouping processing unit 140 generate, for example, a segment definition used for the bucket test executed in the content providing device 30, and transmit it to the content providing device 30. The segment definition is information indicating that a user has been assigned to any of a plurality of segments.

FIG. 4 is a sequence diagram showing an example of the flow of the bucket test process. First, the content providing device 30 transmits a segment definition request for which the number of seeds (the number of layers) and the like is specified, and at least a part of the user information 38 to the information processing device 100 (S1).

The information processing device 100 performs coding processing of the user's identification information based on the designated number of seeds (number of layers) (S2), generates a segment definition based on the result of the coding processing (S3), and then generates a segment definition (S3). It is transmitted to the content providing device 30 (S4). The number of seeds (number of layers) may be specified by the operator of the information processing device 100 using an input device (mouse, keyboard, touch panel, etc.) (not shown) of the information processing device 100.

In the content providing device 30, the distribution setting is different for each segment defined by the received segment definition, the same distribution setting is applied to users belonging to the same segment, and the main content is distributed for a certain period of time (S5). ). Then, the distribution setting in which the indicators such as CTR (Click Through Rate) and CVR (Conversion Rate) related to the advertisement are good within that period is adopted, and it is applied to the subsequent provision of the main content (S6).

Hereinafter, the processing of each part of the information processing apparatus 100 will be described. As described above, the seed receiving unit 120 receives the designation of the number of seeds (the number of layers) from the content providing device 30 or via an input device (not shown).

The coding processing unit 130, for example, inputs the user's identification information into the hash function and calculates the hash value. In the hash function, seeds corresponding to the number of seeds received by the seed receiving unit 120 are input.

The grouping processing unit 140 groups (segments) users based on the hash value calculated by the coding processing unit 130, and generates a segment definition.

FIG. 5 is a diagram schematically showing the content of processing by the coding processing unit 130. The coding processing unit 130 inputs the identification information and the seed for each layer into the hash function, and obtains the hash value for each layer. The coding processing unit 130 calculates a MAC value based on, for example, HMAC (Hash-based Message Authentication Code). The hash function used is, for example, any iterative hash function such as MD5 or SHA-1. The number of layers is n, which matches the number of seeds. n is an arbitrary natural number. The coding processing unit 130 automatically generates n seeds (1) to seeds (n) that are different from each other according to random number processing or a predetermined rule, or accepts and sets an external designation. As a result, the coding processing unit can output a hash value of the number of layers n for each identification information, and the number n of layers can be arbitrarily changed. The hash value for each layer is referred to as a layer (1) hash value to a layer (n) hash value.

FIG. 6 is a diagram schematically showing the processing contents of the grouping processing unit 140. The grouping processing unit 140 determines whether each of the hash values for each layer calculated by the coding processing unit 130 corresponds to, for example, each of the regions h1 to hm in which the range in which the hash value can be taken is arbitrarily divided. Judge and apply to the relevant area. In the example of FIG. 6, the range in which the hash value can be taken is evenly divided into m (m is an arbitrary natural number), the layer (1) hash value is classified into h4, and the layer (2) hash value is classified into h2. And the layer (n) hash value is classified as h15. Hereinafter, the areas in which the hash values of each layer are classified are referred to as segments S (1) to S (n) for each layer.

For example, the grouping processing unit 140 sets each of the segments S (1) to S (n) for each layer as a plurality of segments to which the user related to the identification information belongs at the same time. In this case, one user belongs to each of n segments for each layer, and can be considered to have an attribute represented by an n-dimensional vector.

Further, the grouping processing unit 140 combines (concatenate) the values indicated by the logical products of the classification results between some or all layers (for example, each of the segments S (1) to S (n) for each layer). The thing) may be a segment to which the user belongs. In this case, the user belongs to one of the segments having n × m choices, and can be considered to have an attribute represented by a scalar value.

FIG. 7 is a flowchart showing an example of the flow of processing executed by the information processing apparatus 100. The processing of this flowchart corresponds to the processing of S3 in FIG.

First, the information processing device 100 acquires user information from the content providing device 30 (S11) and accepts the designation of the number of seeds (number of layers) (S12). Next, the information processing apparatus 100 sets seeds (1) to seeds (n) for each layer (S13).

Next, the information processing apparatus 100 executes the processes of S14 and S15 for all the identification information included in the user information acquired in S11 and for all the layers. As described above, the coding processing unit 130 calculates the hash value based on the identification information and the seed for each layer (S14), and the grouping processing unit 140 determines the segment S (i) for each layer related to the identification information. (S15). Note that "i" indicates a layer selected in the process.

When the processes of S14 and S15 are executed for all the identification information and all the layers, the grouping processing unit 140 generates segment definition information based on the segments (for all identification information) of all users (for all identification information). S16), the information is transmitted to the content providing device 30.

The content providing device 30 that has received the segment definition information provides the main content to the user's terminal device 10 with different distribution settings for each segment based on the segment definition information, and sets the distribution settings for which good results are obtained. Adopt and provide content. At this time, the content providing device 30 divides the distribution settings, assigns each divided partial distribution setting to each layer, integrates the partial distribution settings for which suitable results are obtained for each layer, and then integrates the subsequent distribution settings. May be. This will be described below.

FIG. 8 is a diagram schematically showing an example of processing executed by the content providing device 30. In this figure, in order to simplify the explanation, it is assumed that n = 3 and m = 3. In this case, the segments S (1) to S (3) for each layer are classified into any of h1 to h3. Although it is shown that the region division is uniform for each layer, the mode of region division may be different for each layer.

For example, the content providing device 30 allocates the partial distribution setting [number of ad slots] based on the classification of the layer (1), and sets the partial distribution setting [mixed ratio of user interests] based on the classification of the layer (2). Allocation, partial delivery setting [ratio of general-purpose advertisement to targeting advertisement] is assigned based on the classification of layer (3). In this case, assuming that the classification for which the most suitable result was obtained was h2 (1) in the layer (1), h1 (2) in the layer (2), and h3 (3) in the layer (3), the content. The providing device 30 includes the number of ad slots assigned to the classification h2 (1), the mixed ratio of interests assigned to the classification h1 (2), and the general-purpose advertisement and the targeting advertisement assigned to the classification h3 (3). The ratio of is adopted and set as the delivery setting. As a result, the bucket test can be performed in a multifaceted and complex manner, and the test accuracy can be improved.

Here, a comparison between the comparative example and the embodiment will be described. The information processing device A in the comparative example does not obtain a hash value for each layer, but obtains one hash value from user identification information or the like based on a certain function, and determines a user segment based on the hash value. It is a thing. In the information processing apparatus A of the comparative example, the segment definition is the same for each bucket test, and a different segment definition cannot be generated for each bucket test. Therefore, since the users belonging to each segment are the same users each time, the tendency of the users is fixed, and the test result may be unnecessarily biased. In addition, when the bucket test is performed multiple times, an appropriate result may not be obtained due to a dependency relationship with the past test results.

In this respect, the information processing apparatus 100 of the embodiment accepts the designation of the number of seeds, and the seeds themselves can be arbitrarily set each time, so that the bucket test can be performed flexibly and with little bias in the results. On the contrary, if you want to create a dependency with the past test results, you can fix the number of seeds and the seeds.

Further, since the information processing apparatus 100 of the embodiment determines a segment for each layer, when it is desired to perform a bucket test by changing various distribution settings, the bucket test is performed without significantly reducing the number of users in the segment. Can be done. If the user group is simply divided by a large number of divisions, the number of users in the segment will be reduced, but it is possible to prevent such a situation from occurring.

Further, the information processing apparatus 100 of the embodiment generates a segment definition according to the number of distribution setting patterns on the content providing apparatus 30 side by arbitrarily changing the number of seeds and the number of divisions of possible values of the hash value. can do. That is, it is possible to generate a segment definition that matches the circumstances of the user of the segment definition.

As described above, the information processing device 100 may be a function of the content providing device 30 or the advertisement distribution device 50, that is, a virtual device. In this case, the information processing device 100 may be the web management unit 32 or the advertisement. It may include a determination unit 34.

According to the information processing apparatus 100 of the embodiment described above, the coding processing unit 130 that performs coding processing (hashing processing) on the identification information that identifies the user who browses the main content, and the coding processing are performed. A group that classifies the obtained results (hash values), groups the users (assigns them to segments), and outputs information (segment definitions) of the user groups that share the same settings related to the main content based on the grouped results. The coding processing unit 140 includes a coding processing unit 140, and the coding processing unit 130 performs coding processing on the same identification information group in a plurality of layers by changing the coding processing parameter (seed), and groups the same information group. Since the 140 groups the users for each layer or groups the users based on the logical product of the classification results between the layers, it is possible to flexibly define a group of users who have common settings related to the content.

Although the embodiments for carrying out the present invention have been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. Can be added.

10 Terminal device 30 Content providing device 32 Web management unit 34 Advertising decision unit 36 Main content 38 User information 50 Advertising distribution device 52 Advertising content 100 Information processing device 110 Communication unit 120 Seed reception unit 130 Coding processing unit 140 Grouping processing unit NW network

Claims (8)

A coding processing unit that performs coding processing on the identification information that identifies the user who browses the content,
A grouping processing unit that classifies the results of the coding processing, groups the users, and outputs information on a group of users who share the same settings related to the content based on the grouping results.
With
The coding processing unit performs coding processing on the same identification information group in a plurality of layers by changing the parameters of the coding processing.
The grouping processing unit groups the users for each layer, or groups the users based on the logical product of the classification results between the layers.
Information processing device. The setting related to the content is the distribution setting of the advertisement embedded in the content.
The information processing device according to claim 1. The coding processing unit performs the input of the identification information and the seed, which is the parameter, into the hash function as the coding process.
The information processing device according to claim 1 or 2. The coding processing unit can change the number of the layers based on the input information.
The information processing device according to any one of claims 1 to 3. Based on the information of a group of users who share the same content settings, the contents are provided to the users with different settings for each group, and the contents are provided by adopting the settings for which good results are obtained. Further equipped with a distribution unit
The information processing device according to any one of claims 1 to 4. The distribution unit integrates the settings for which good results have been obtained for each layer, and sets the settings when providing the content.
The information processing device according to claim 5. The computer
Encoding processing is performed on the identification information that identifies the user who browses the content.
Users are grouped by classifying the results of the coding process.
An information processing method that outputs information on a group of users who share settings related to the content based on the grouping result.
When performing the coding process, the same identification information group is coded in a plurality of layers by changing the parameters.
At the time of grouping, the users are grouped for each layer, or the users are grouped based on the logical product of the classification results between the layers.
Information processing method. On the computer
Encoding processing is performed on the identification information that identifies the user who browses the content.
The results of the coding process are classified and the users are grouped.
A program that outputs information on a group of users who share settings related to the content based on the grouping result.
On the computer
When the coding process is performed, the same identification information group is subjected to the coding process in a plurality of layers by changing the parameters.
At the time of grouping, the users are grouped for each layer, or the users are grouped based on the logical product of the classification results between the layers.
program.

Images