JP2017204134A - Attribute estimation device, attribute estimation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate a user's psychographic attribute without analyzing in advance information on the user.SOLUTION: An attribute estimation device 13 according to the present invention estimates a psychographic attribute indicating a user's inner characteristics, and comprises: a content control part 133 that presents a content corresponding to the psychographic attribute; and an attribute estimation part 134 that estimates if the user belongs to a specific psychographic attribute on the basis of the user's reaction to the presented content.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an attribute estimation apparatus, an attribute estimation method, and a program for estimating a psychographic attribute indicating an internal feature of a user.

  In a place where a large number of people gather, such as airports, train stations, commercial facilities, etc., for the purpose of guidance and advertisement to the user, the content for guidance and advertisement to the user is presented using a projector or a display device. Digital signage is popular. In order to enhance the effect (for example, advertisement effect) of content presentation by digital signage, it is important to present content corresponding to the attribute of the user viewing the content.

  The user attributes include demographic attributes indicating appearance characteristics such as age and sex, and psychographic attributes indicating internal characteristics such as ethnicity, nationality, language used, and preferences.

  As a technique for estimating user attributes, Non-Patent Document 1 shoots a user who exists in an advertisement-visible area, and estimates attributes (demographic attributes) such as the user's sex and age from the captured image. Techniques to do this are disclosed.

  Patent Document 1 discloses a technique for estimating a user attribute (psychographic attribute) by analyzing contents such as a diary or a profile posted by a user to an SNS (Social Networking Service).

JP 2013-218579 A

Hiroyuki Arai, Naomi Ito, Kaori Kataoka, Yukinobu Taniguchi, “Advertising Effect Measurement Technology Using Image Processing”, NTT Technical Journal, January 2013, p.61-64, [online], [March 4, 2016 Search], Internet <URL: http://www.ntt.co.jp/journal/1301/files/jn201301061.pdf>

  In digital signage, it is important to present not only the demographic attribute of the user but also the content corresponding to the psychographic attribute of the user in order to increase the effect of presenting the content (for example, the advertising effect).

  In the technique disclosed in Non-Patent Document 1, only demographic attributes can be estimated, and psychographic attributes can be estimated in order to estimate user attributes from user-captured images existing in a viewable area of an advertisement. I can't do it.

  In the technique disclosed in Patent Document 1, it is necessary to analyze information about the user such as a diary or a profile posted by the user on the SNS. If there is no such information, the psychographic attribute of the user is estimated. I can't do it. In particular, since digital signage targets a large number of unspecified users, it is difficult to analyze information about users in advance, and the technique disclosed in Patent Document 1 is used to estimate user attributes in digital signage. It is difficult to apply.

  An object of the present invention made in view of the above problems is to provide an attribute estimation device, an attribute estimation method, and a program capable of estimating a psychographic attribute of a user without performing prior analysis of information about the user. It is to provide.

  In order to solve the above problems, an attribute estimation apparatus according to the present invention is an attribute estimation apparatus that estimates a psychographic attribute indicating an internal feature of a user, and presents content corresponding to a specific psychographic attribute A control unit; and an attribute estimation unit configured to estimate whether the user belongs to the specific psychographic attribute based on a user's reaction to the presented content.

  In order to solve the above problem, an attribute estimation method according to the present invention is an attribute estimation method by an attribute estimation apparatus that estimates psychographic attributes indicating an internal feature of a user. Presenting content corresponding to a psychographic attribute, and an attribute estimating unit estimating whether the user belongs to the specific psychographic attribute based on a user's reaction to the presented content. .

  In order to solve the above problems, the program according to the present invention presents content corresponding to a specific psychographic attribute to a computer in the attribute estimation apparatus that estimates a psychographic attribute indicating the internal characteristics of the user. A process and a process of estimating whether the user belongs to the specific psychographic attribute based on a user's reaction to the presented content.

  According to the attribute estimation device, the attribute estimation method, and the program according to the present invention, it is possible to estimate a user's psychographic attribute without performing prior analysis of information about the user.

It is a figure which shows the principal part structure of the attribute estimation system which concerns on one Embodiment of this invention. It is a figure which shows an example of a structure of the content data which the content data storage part shown in FIG. 1 memorize | stores. It is a flowchart which shows an example of the operation | movement at the time of the user's attribute estimation of the signage server shown in FIG. It is a figure which shows an example of a structure of the attribute estimation data which the attribute estimation part shown in FIG. 1 manages. It is a figure which shows an example of the operation | movement at the time of content control of the signage server shown in FIG.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. It should be noted that the present invention is not limited to the following embodiments, and various modifications can be made without departing from the spirit of the present invention.

  FIG. 1 is a diagram showing a main configuration of an attribute estimation system 10 according to an embodiment of the present invention. The attribute estimation system 10 according to this embodiment uses a projector, a display device, or the like to present content for guidance, advertisement, etc. to the user, and to display psychographic attributes indicating the internal characteristics of the user. To be estimated.

  The attribute estimation system 10 shown in FIG. 1 includes a plurality of signage presentation units 11 (11-1, 11-2,..., 11-n) and a plurality of reaction observation units 12 (12-1, 12-2, .., 12-n) and a signage server 13 (attribute estimation device).

  Signage presentation units 11-1, 11-2,... 11-n are installed in an area (for example, a facility) where digital signage guidance, advertisements, and the like are provided. The signage presentation unit 11 presents content for guidance, advertisement, etc. to the user according to the control of the signage server 13. Specific examples of the signage presentation unit 11 include a display device that displays content on a display surface and a projector that projects content on a projection surface. The content format is not limited to a specific format. The content includes not only jpg and gif format images and still images, but also mp4 format moving images, HTML format content, JavaScript (registered trademark) and libraries (such as canvas in HTML5). Various formats are included.

  The reaction observation units 12-1, 12-2,..., 12-n are provided corresponding to the signage presentation units 11-1, 11-2,. The reaction observation unit 12 is a camera or the like that captures an area where the content presented by the corresponding signage presentation unit 11 can be visually recognized, and outputs the captured image to the signage server 13.

  The signage presentation unit 11 and the reaction observation unit 12 do not need to correspond one-to-one. For example, when the visible area of the content presented by the signage presenting unit 11 is narrow and the plurality of signage presenting units 11 are arranged at a short distance, the plurality of signage is displayed by one reaction observation unit 12. An area where the content presented by the presentation unit 11 is visible may be photographed. In addition, when the visible area of the content presented by one signage presenting unit 11 is wide, the visible area of the content presented by the one signage presenting unit 11 is photographed by a plurality of reaction observing units 12. Also good.

  The signage server 13 causes the signage presentation unit 11 to present content. Further, the signage server 13 detects a user reaction from a captured image of the reaction observation unit 12 corresponding to the signage presentation unit 11 that presents the content, and estimates an attribute (psychographic attribute) of the user who views the content. The estimation result of the user attribute is used for presenting content corresponding to the user attribute. As a user's reaction detected from the captured image of the reaction observation unit 12, for example, there is a reaction of whether or not the user actually took off the shoes with respect to the content that prompts the user to take off the shoes.

  The detection of the reaction of the user who sees the content is not limited to the detection by the captured image of the reaction observation unit 12. For example, when detecting a user's movement, it is also possible to detect from the communication log of the communication apparatus (for example, smart phone etc.) and the communication equipment in a facility which a user carries. For example, when the unique address of the communication device owned by the user is detected by the second communication facility from the first communication facility, the user receives the second communication from the vicinity of the first communication facility. It can be estimated that it has moved to the vicinity of the equipment. For example, a communication facility that covers an area where the content presented by a signage presentation unit 11 can be visually recognized is a first communication facility, and a communication facility that covers a guided area is a second communication facility. As a result of the user reacting to the sign, it can be estimated that the user has moved from the vicinity of the first communication facility to the vicinity of the second communication facility.

  Next, the configuration of the signage server 13 will be described. In addition, since the structure of the signage presentation part 11 and the reaction observation part 12 is well known to those skilled in the art, and since it is not directly related to this invention, description is abbreviate | omitted.

  The signage server 13 illustrated in FIG. 1 includes a content data storage unit 131, a communication unit 132, a content control unit 133, and an attribute estimation unit 134.

  The content data storage unit 131 stores content and attributes (for example, psychographic attributes) corresponding to the content as content data.

  FIG. 2 is a diagram illustrating an example of the configuration of content data.

  As shown in FIG. 2, for each content ID (content ID), the content data storage unit 131 includes an attribute ID indicating an attribute corresponding to the content, and when the user who viewed the content reacts to the content, The estimation accuracy (probability) that the user belongs to the attribute corresponding to the content is stored.

  Referring back to FIG. 1, the communication unit 132 communicates with the signage presentation units 11-1, 11-2,... 11 -n, and transmits content to be presented to each signage presentation unit 11, for example. In addition, the communication unit 132 communicates with the reaction observation units 12-1, 12-2,..., 12-n, and receives a captured image from each reaction observation unit 12, for example.

  The content control unit 133 causes the signage presentation unit 11 to present the content stored in the content data storage unit 131 via the communication unit 132. Further, the content control unit 133 outputs the content data of the presented content to the attribute estimation unit 134.

  The attribute estimation unit 134 acquires a captured image via the communication unit 132 from the reaction observation unit 12 corresponding to the signage presentation unit 11 that presents the content. And the attribute estimation part 134 detects a user's reaction with respect to the content shown to the signage presentation part 11 from the picked-up image of the reaction observation part 12, and based on the detection result, the attribute corresponding to the content presented by the user To belong to.

  Next, the operation of the signage server 13 according to the present embodiment will be described.

  FIG. 3 is a flowchart illustrating an example of an operation when the signage server 13 estimates a user attribute (psychographic attribute).

  First, the content control unit 133 causes the signage presentation unit 11 to present content corresponding to a specific psychographic attribute (step S11). As such content, for example, for a user belonging to a specific psychographic attribute (a user who uses Chinese as a language), such as a content indicating an action guide such as “the north exit is over there” in Chinese. Content known to work on is used.

  The attribute estimation unit 134 acquires a captured image from the reaction observation unit 12 corresponding to the signage presentation unit 11 that presents content. Then, the attribute estimation unit 134 determines whether or not the user present in the area where the content presented to the signage presentation unit 11 can be visually recognized from the captured image of the reaction observation unit 12 has reacted to the presented content. Is determined (step S12).

  As a user's reaction to the content, the user has seen the content for a predetermined time or more, and the user has acted according to the behavior guidance indicated in the content. Such a user's reaction can be detected by performing an appearance observation of the user by image analysis of the captured image of the reaction observation unit 12. Whether or not the user has viewed the content can be detected using any gaze detection technique. Further, whether or not the user has acted along the behavior guidance indicated in the content can be detected by using any behavior detection technology (for example, the technology using the communication log described above). The line-of-sight detection technique and the action detection technique as described above are well known to those skilled in the art and are not directly related to the present invention, and thus the description thereof is omitted.

  When it is determined that the user does not react to the presented content (step S12: No), the attribute estimation unit 134 estimates that the user does not belong to the attribute corresponding to the presented content. Then, for example, the process returns to step S11, and content corresponding to another attribute is presented. In addition, operation | movement when a user does not react with respect to the shown content is not restricted to this, For example, the signage server 13 may complete | finish a process.

  If it is determined that the user has reacted to the presented content (step S12: Yes), the attribute estimation unit 134 estimates that the user belongs to the attribute corresponding to the presented content. Here, the attribute estimation unit 134 obtains the result of user attribute estimation as the accuracy of estimation that the attribute belongs. FIG. 4 is a diagram illustrating an example of a configuration of attribute estimation data indicating a result of user attribute estimation managed by the attribute estimation unit 134.

  As shown in FIG. 4, the attribute estimation unit 134 manages, for each user ID, an attribute ID indicating an attribute to which the user belongs and an estimation accuracy that the attribute belongs to the attribute estimation data.

  When the attribute estimation unit 134 presents content corresponding to a specific psychographic attribute and determines that the user has reacted to the presented content, the attribute estimation unit 134 assigns a user ID to the user, and the user assigns the specific psychographic attribute to the user. The estimation accuracy of belonging is assigned (step S13). The estimation accuracy is given based on the content data output from the content control unit 133, for example. The attribute estimation unit 134 is included in the captured image of the reaction observing unit 12, and gives an estimation accuracy that it belongs to a psychographic attribute corresponding to the content for each user who has reacted to the content.

  By repeating the presentation of the content and the user attribute estimation, the psychographic attribute of the user can be estimated with high accuracy without prior analysis of information related to the user. Further, the attribute estimation unit 134 can increase the estimation accuracy by various methods.

  For example, the attribute estimating unit 134 causes the signage presenting unit 11 to present a plurality of contents corresponding to the same psychographic attribute and having different contents, and when the user reacts to each of the plurality of contents, the plurality of contents presented by the user are displayed. The accuracy of estimation of belonging to a psychographic attribute corresponding to the content may be increased. Such a plurality of contents may be presented at the same place (same signage presentation unit 11) or may be presented at different places (different signage presentation units 11).

  In addition, when the user reacts to content corresponding to a specific psychographic attribute, the attribute estimation unit 134 determines that the user has reacted to the content (behaved according to the content) as a result of the behavior detection. The accuracy of estimation that a user belongs to a specific psychographic attribute may be increased. Furthermore, the attribute estimation unit 134 performs gaze detection together with behavior detection, and when it is determined that the user has responded to the content (behaved according to the content) after viewing the content, the user has set a specific psychographic attribute. The accuracy of estimation of belonging may be further increased.

  For example, at an airport, it is estimated that there are more Chinese people at check-in counters for Chinese flights than at check-in counters for flights to other countries. When a user who comes from such an area reacts by seeing content corresponding to Chinese, the nationality of the user is likely to be Chinese.

  Therefore, the attribute estimation unit 134, when a user who comes from an area where it is estimated that users belonging to a specific psychographic attribute are concentrated, reacts to content corresponding to the specific psychographic attribute, The accuracy of estimation of belonging to a specific psychographic attribute may be higher than the accuracy with respect to users coming from other areas.

  Further, the accuracy of estimation may be changed according to the detection technique used for detecting the user's reaction. For example, when the method using image analysis is more accurate than the method using communication logs, the accuracy to be added to the accuracy of estimation based on image analysis may be increased.

  In addition, as content presented at the time of user attribute estimation, it is possible to use content indicating an advertisement (commercial advertisement, event notification, etc.) to the user. However, as described above, action guidance to the user is provided. By using the content to be shown, more accurate estimation is possible. Unlike the case where the content showing the advertisement to the user is presented, when the content showing the behavior guidance to the user is presented, the user views the content if the behavior guidance indicated by the content is useful for the user. Later, he often takes action according to the action guide promptly. Therefore, a user who has acted promptly according to the behavior guidance indicated by the content is a user who understands the content of the behavior guidance indicated by the content and recognizes the content as the behavior guidance addressed to himself, that is, corresponds to the content. It can be estimated with high accuracy that the user belongs to the attribute.

  The signage server 13 may control the content to be presented to the signage presentation unit 11 based on the result of the user attribute estimation by the attribute estimation unit 134. FIG. 5 is a flowchart showing an operation at the time of content control by the signage server 13.

  The content control unit 133 presents content corresponding to a certain attribute (psychographic attribute) to the signage presenting unit 11 (step S21). The attribute estimation unit 134 acquires a captured image from the reaction observation unit 12 corresponding to the signage presentation unit 11 that presents the content, and counts the number of users who have reacted to the presented content from the acquired captured image (step S22). ). The content control unit 133 and the attribute estimation unit 134 repeat the above-described processing for a plurality of attributes.

  Next, the content control unit 133 causes the signage presentation unit 11 to present content corresponding to the attribute having the highest ratio of the number of users who have reacted to the content among the plurality of attributes (step S23). The attribute with the highest ratio of the number of users who react to the content is considered to be the attribute to which the user with the highest ratio of the numbers belongs in the area where the content is presented. Therefore, the content presentation effect can be enhanced by presenting the content corresponding to the attribute of the user having a high number ratio.

  As described above, according to the present embodiment, the signage server 13 includes a content control unit 133 that presents content corresponding to a specific psychographic attribute and a user's specific psychographic based on the user's reaction to the presented content. And an attribute estimation unit 134 that estimates whether the attribute belongs.

  When the content corresponding to a specific psychographic attribute is presented and the user reacts to the presented content, the user is likely to understand the content and recognize the content as intended . That is, a user who reacts to the presented content is likely to belong to the psychographic attribute. Therefore, the user's psychographic attribute can be estimated from the user's reaction to the content corresponding to the psychographic attribute without analyzing information about the user.

  In the present embodiment, the content corresponding to a specific psychographic attribute is presented, and the user's psychographic attribute is estimated only by observing the user's reaction to the presented content. Therefore, a system can be constructed relatively easily using an existing device such as a projector or a display device as the signage presentation unit 11 and a camera as the reaction observation unit 12. Therefore, estimating the psychographic attribute of the user without using a special device for collecting information related to the user (content such as a diary or profile posted on the SNS) as disclosed in Patent Document 1. Can do.

  In order to make the website more effective (increase the number of accesses), web pages with different patterns are prepared, and the web pages with the patterns are actually used by the user for access analysis. Thus, a test (A / B test) for comparing effects may be performed. Content presented in digital signage can also be made more effective (for example, higher inductive effect) by the same method as the A / B test.

  For example, for a certain content, a plurality of contents are prepared by changing some of the constituent elements such as sentences and symbols constituting the content. Then, a plurality of prepared contents can be respectively presented, and it is possible to specify what kind of components are effective from the user's reaction to each content. Also, unlike website access analysis, what type of content is effective for users belonging to a certain attribute by estimating the user's attributes (psychographic attributes) for each content. It is possible to grasp the trend. Furthermore, the accuracy of attribute estimation can be improved by utilizing such a tendency for user attribute estimation.

  By using the present invention, it is also possible to determine whether the content corresponds to a specific psychographic attribute (acts on a user belonging to the specific psychographic attribute). Specifically, by presenting content that is known to correspond to a specific psychographic attribute, estimating users belonging to the specific psychographic attribute, and presenting other content to the user, It is also possible to determine whether or not the content corresponds to a specific psychographic attribute to which the user belongs. Also, for example, when using digital signage at a new facility, it is difficult to know in advance what kind of information will be given at what point (what kind of information is needed at which point). is there. Therefore, a plurality of contents indicating a plurality of pieces of information (messages) assumed to be required at a certain point are prepared, and the plurality of contents are presented at the point. Then, by detecting how the user reacts to each content, for example, to which content the user reacts most, the information required at that point is grasped. It is also possible.

  The method performed in the signage server 13 according to the present invention may be applied to a program for causing a computer to execute. In addition, the program can be stored in a storage medium and can be provided to the outside via a network.

  Although the present invention has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various variations or modifications based on the present disclosure. Therefore, it should be noted that these variations or modifications are included in the scope of the present invention.

DESCRIPTION OF SYMBOLS 10 Image processing system 11 Signage presentation part 12 Reaction observation part 13 Signage server 131 Content data storage part 132 Communication part 133 Content control part 134 Attribute estimation part

Claims (8)

An attribute estimation device for estimating psychographic attributes indicating an internal feature of a user,
A content control unit that presents content corresponding to a specific psychographic attribute;
An attribute estimation unit that estimates whether the user belongs to the specific psychographic attribute based on a user's reaction to the presented content;
An attribute estimation device comprising: The attribute estimation apparatus according to claim 1,
The attribute estimation unit obtains the accuracy of estimation that the user belongs to the specific psychographic attribute. The attribute estimation apparatus according to claim 2,
The content control unit corresponds to the same psychographic attribute and presents a plurality of different contents to the user,
The attribute estimation unit increases the accuracy of estimation that the user belongs to the specific psychographic attribute when the user reacts to each of the plurality of contents. The attribute estimation device according to claim 2 or 3,
The attribute estimation unit increases the accuracy of estimation that the user belongs to the specific psychographic attribute when it is determined that the user has acted according to the content based on a result of the user's behavior detection. Attribute estimation device. The attribute estimation device according to claim 4,
The attribute estimation unit further detects the user's line of sight, and based on the results of the user's line of sight detection and the user's action detection, the user behaves according to the content after viewing the presented content The attribute estimation device further increases the accuracy of estimation that the user belongs to the specific psychographic attribute. The attribute estimation device according to any one of claims 2 to 5,
The attribute estimation unit may be configured such that when a user who comes from an area where it is estimated that users belonging to a specific psychographic attribute are concentrated and reacts to content corresponding to the specific psychographic attribute, the user An attribute estimation apparatus characterized by increasing the accuracy of estimation of belonging to a psychographic attribute of An attribute estimation method by an attribute estimation device that estimates psychographic attributes indicating an internal feature of a user,
A content control unit presenting content corresponding to a specific psychographic attribute;
An attribute estimation unit including: estimating whether the user belongs to the specific psychographic attribute based on a user response to the presented content. In a computer in the attribute estimation device that estimates psychographic attributes indicating the internal characteristics of the user,
Processing to present content corresponding to specific psychographic attributes;
A program for executing a process of estimating whether the user belongs to the specific psychographic attribute based on a user's reaction to the presented content.

Images