Information processing program, terminal device, and information processing method

Abstract

To deliver appropriate advertisements to users while protecting their personal information. [Solution] The information processing program according to the present invention causes the computer of a terminal device to execute a generation procedure that generates an advertisement for a specified target using user information held by the terminal device and an advertisement generation model that has been trained to generate an advertisement for a specified target; an evaluation procedure that evaluates the advertisement generated by the generation procedure using an advertisement evaluation model that has been trained to estimate the evaluation of the advertisement by the user of the terminal device; and a provision procedure that provides the advertisement generated by the generation procedure if the evaluation result from the evaluation procedure satisfies predetermined conditions.

Description

Technical Field

[0001] The present invention relates to an information processing program, a terminal device, and an information processing method.

Background Art

[0002] Conventionally, advertisement distribution via the Internet has been widely performed. As a technology related to advertisement distribution, a technology for distributing advertisements tailored to a user from user log information accumulated in a server has been disclosed (see, for example, Patent Document 1).

Prior Art Literature

Patent Literature

[0003]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0004] However, in the conventional technology, there is a problem that it is difficult to appropriately distribute advertisements to users. In recent years, from the viewpoint of protecting personal information, restrictions are being imposed on user log information provided from user terminals to servers, making it difficult for the server side to analyze what kind of advertisements are highly valuable to users.

[0005] The present invention has been made in view of the above, and an object thereof is to provide an information processing program, a terminal device, and an information processing method capable of appropriately distributing advertisements to users while protecting personal information.

Means for Solving the Problem

[0006] To solve the above-mentioned problems and achieve the objective, the information processing program according to the present invention causes the computer of a terminal device to execute a generation procedure for generating an advertisement for a specified target using user information held by the terminal device and an advertisement generation model that has been trained to generate an advertisement for a specified target; an evaluation procedure for evaluating the advertisement generated by the generation procedure using an advertisement evaluation model that has been trained to estimate the evaluation of the advertisement by the user of the terminal device; and a provision procedure for providing the advertisement generated by the generation procedure if the evaluation result from the evaluation procedure satisfies predetermined conditions. [Effects of the Invention]

[0007] According to the present invention, it is possible to deliver appropriate advertisements to users while protecting personal information. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of information processing according to the embodiment. [Figure 2] Figure 2 shows an example of on-device learning according to the embodiment. [Figure 3] Figure 3 is a block diagram showing an example configuration of an information providing device according to an embodiment. [Figure 4] Figure 4 shows an example of an advertising information database according to the present invention. [Figure 5] Figure 5 shows an example of a statistical information database according to the embodiment. [Figure 6] Figure 6 is a block diagram showing an example configuration of a user terminal according to the embodiment. [Figure 7] Figure 7 shows an example of a log information storage unit according to the embodiment. [Figure 8] Figure 8 shows an example of a camera image storage unit according to an embodiment. [Figure 9] Figure 9 shows an example of an advertising information storage unit according to the embodiment. [Figure 10]Figure 10 is a schematic diagram showing an example of advertisement generation using camera images according to the embodiment. [Figure 11] Figure 11 is a schematic diagram showing an example of providing advertisements according to the surrounding environment according to the embodiment. [Figure 12] Figure 12 is a flowchart showing an example of a processing procedure performed by the information providing device according to this embodiment. [Figure 13] Figure 13 is a flowchart showing an example of a processing procedure performed by a user terminal according to this embodiment. [Figure 14] Figure 14 is a flowchart showing an example of a processing procedure performed by a user terminal according to this embodiment. [Figure 15] Figure 15 is a flowchart showing an example of a processing procedure performed by a user terminal according to this embodiment. [Figure 16] Figure 16 is a flowchart showing an example of a processing procedure executed by a user terminal according to the embodiment. [Figure 17] Figure 17 is a hardware configuration diagram showing an example of a computer that implements the functions of a user terminal according to the embodiment. [Modes for carrying out the invention]

[0009] The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing program, terminal device, and information processing method according to the present application (hereinafter referred to as "embodiments"). Note that these embodiments do not limit the information processing program, terminal device, and information processing method according to the present application. Furthermore, each embodiment can be combined as appropriate, provided that the processing content is not inconsistent. Also, the same parts are denoted by the same reference numerals in each of the following embodiments, and redundant descriptions are omitted.

[0010] [1. Overview of Information Processing] Figure 1 is a diagram showing an example of information processing according to the embodiment, and the information processing program according to the embodiment is executed by the user terminal 10 shown in Figure 1.

[0011] The advertisement distribution system S illustrated in FIG. 1 includes an information providing device 1, a user terminal 10, and advertiser terminals 100a, 100b, 100c. The information providing device 1 provides advertisement information related to the requested advertisement to each user terminal 10 based on advertisement distribution requests received from the advertiser terminals 100a, 100b, 100c. For example, the information providing device 1 is implemented by a cloud, a server, or the like.

[0012] Furthermore, the user terminal 10 illustrated in FIG. 1 is a terminal device owned by a user U, and is, for example, a smartphone or the like. As illustrated in FIG. 1, the user terminal 10 includes an advertisement generation model and an advertisement evaluation model. The advertisement generation model is a model that has been trained to generate advertisements for a specified target, and is an image generation AI. For example, the advertisement generation model generates an AI image corresponding to input text information or the like (for example, context). Furthermore, the advertisement evaluation model is a model that has been trained to estimate the advertisement evaluation by the user U.

[0013] The user terminal 10 repeatedly executes both processes related to advertisement generation and advertisement evaluation within the user terminal 10. Thereby, the user terminal 10 can optimize the advertisement provided to the user U.

[0014] In other words, the user terminal 10 can provide advertisements optimized for the user U through on-device learning. A specific description of on-device learning will be given later with reference to FIG. 2.

[0015] The advertiser terminals 100a, 100b, 100c are terminal devices of respective advertisers. For example, an advertiser operates each of the advertiser terminals 100a, 100b, 100c to make an advertisement distribution request to the information providing device 1 (step S1).

[0016] Here, the distribution request includes information about the target of the advertisement, material information to be used as advertising material, and trend information regarding advertising trends. Material information includes company name, product name, logo image, product image, etc., and trend information includes information about the image and atmosphere of the advertisement, etc. Furthermore, as will be described later, the user terminal 10 can also generate advertisements using camera images, etc., and in this case, the distribution request includes trend information indicating the trend of images to be combined with the target of the advertisement.

[0017] The information provider 1 provides advertising information to the user terminal 10 based on the distribution request obtained from each advertiser (step S2).

[0018] The advertising information includes, for example, information about materials and advertising trends. The user terminal 10 generates an advertisement using an advertising generation model based on the advertising information received from the information providing device 1 (step S3).

[0019] Next, the user terminal 10 evaluates the generated advertisement using an advertisement evaluation model (step S4). For example, the advertisement evaluation model is a model that learns a person model of user U, including tendencies, type, advertisement preferences, and effectiveness trends, using various user information held within the user terminal 10. The model is trained to give higher scores to advertisements that are estimated to be highly appealing to user U.

[0020] Furthermore, the ad evaluation model can also evaluate the ad provided to user U based on, for example, the relationship between user information before and after ad delivery. The ad evaluation model can also update its score according to, for example, the surrounding environment of user terminal 10. For instance, if the ambient temperature is high, the ad evaluation model might update the score for the advertised beer to a relatively higher value.

[0021] Then, the user terminal 10 provides an advertisement (AD) to the user U when, for example, the advertising score exceeds a predetermined value (step S5). For example, as shown in Figure 1, the advertisement (AD) includes an AI image G and text T generated by the advertisement generation model, and a material image P requested by the advertiser. The advertisement (AD) may be in image format or video format. The advertisement (AD) may also include music. Furthermore, the advertisement (AD) may be three-dimensional content, such as AR (Augmented Reality) content or VR (Virtual Reality) content.

[0022] As will be described later, the user terminal 10 can also generate an advertising ad (AD) that includes a camera image taken by the user terminal 10 and provide it to the user U. In this case, since each user terminal 10 holds a different camera image, each user terminal 10 can provide an advertising ad using a different camera image.

[0023] In particular, when providing advertising ads using camera images, it is possible to provide advertising ads that are linked to the user's past experiences and memories, thus providing advertising ads that are more relatable to the user.

[0024] Subsequently, the user terminal 10 measures the performance of the ad using log information such as the user U's behavioral history acquired after the ad is provided (step S6). The user terminal 10 measures the performance of the ad according to the changes in the user U's behavior towards the ad before and after the ad is provided.

[0025] Subsequently, the user terminal 10 provides the information provider 1 with statistical information necessary for the information provider 1 to bill the advertiser for costs, etc. (Step S7). For example, the statistical information is minimal, such as information on the features of the ad and the results of performance measurement, and does not include sensitive information that would identify the user U personally.

[0026] In other words, the user terminal 10 does not provide sensitive information, such as user U's log information, to the information provision device 1. This protects user U's personal information. Note that the features of the ad are feature vectors that represent the features of the ad, but they may also be text that represents the features of the ad, for example.

[0027] The information provider 1 generates a report by aggregating statistical information acquired from each user terminal 10 (step S8), and provides the generated report to each advertiser terminal 100a, 100b, and 100c (step S9).

[0028] For example, the report includes information such as the distribution of features and centroids for each ad, and the distribution of performance scores for each ad. This allows advertisers to understand the ad provided on each user device 10 and the performance status of the ad.

[0029] Next, we will further explain on-device learning using Figure 2. Figure 2 is a schematic diagram showing an example of on-device learning according to the embodiment. As shown in Figure 2, the user terminal 10 places an advertisement order from the advertisement evaluation model to the advertisement generation model (step S11).

[0030] The ad order includes information about the characteristics of an ad that is estimated to be highly appealing to user U, and the ad generation model generates an ad based on the ad order (step S12). The user terminal 10 provides the ad generated using the ad generation model to user U at a predetermined time (step S13).

[0031] Subsequently, the user terminal 10 collects log information from user U (step S14). Then, the user terminal 10 evaluates the ad (AD) based on the collected log information (step S15). The user terminal 10 may also evaluate the ad (AD) using methods such as A / B testing or CVR (Conversion Rate).

[0032] Then, the user terminal 10 improves the order based on the evaluation results of the advertisement AD provided to user U. In other words, by repeating the loop of steps S11 to S15 within the user terminal 10, it is possible to provide appropriate advertisements to user U without uploading personal information about user U to a server (for example, information provision device 1).

[0033] Therefore, according to the user terminal 10 of this embodiment, it is possible to deliver appropriate advertisements to users while protecting personal information. The advertisement creation model and the advertisement evaluation model may be, for example, Generative Adversarial Networks (GANs). In this case, the advertisement creation model may be modified based on the evaluation results of the advertisement evaluation model, and the advertisement evaluation model may be modified based on the creation results of the advertisement creation model. For example, the advertisement creation model may be modified so that the evaluation results of the advertisement evaluation model are better, and the advertisement evaluation model may be modified so that the content generated by the advertisement creation model is evaluated more strictly. For example, in this case, the advertisement evaluation model is modified so that an advertisement that user U rated as good is given a lower score than the previous time, so that the evaluation more strongly reflects user U's preferences.

[0034] [2. Example of information provision device configuration] Next, an example of the configuration of the information providing device 1 according to the embodiment will be described using Figure 3. Figure 2 is a block diagram showing an example of the configuration of the information providing device 1 according to the embodiment. As shown in Figure 2, the information providing device 1 according to the embodiment includes a communication unit 2, a storage unit 3, and a control unit 4.

[0035] The communication unit 2 is implemented, for example, by a NIC (Network Interface Card). The communication unit 2 transmits and receives information with external devices via various wireless communication networks such as 4G (Generation), 5G, LTE (Long Term Evolution), WiFi (registered trademark), or wireless LAN (Local Area Network), or various wired communication networks.

[0036] The storage unit 3 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, or storage devices such as hard disks and optical discs. The storage unit 3 also includes an advertising information database 31, a statistical information database 32, and an estimation model database 33.

[0037] The advertising information database 31 is a database that stores advertising information. Figure 4 shows an example of the advertising information database 31 according to this embodiment. As shown in Figure 4, the advertising information database 31 stores information on items such as "advertiser ID," "advertising target," "material information," "trend information," and "target information" in association with each other.

[0038] "Advertiser ID" is an identifier used to identify the advertiser. "Target" refers to the store, brand, event, product, or service that is the target of the advertisement. "Material Information" is information about the materials used in the advertisement. For example, material information may include product images, the advertiser's logo or character images, images of celebrities featured in the advertisement, and music used in the advertisement.

[0039] "Trend information" refers to information about the trends of the advertisement as specified by the advertiser. For example, trend information includes the image the advertiser wants from the advertisement and the advertiser's brand image. "Target information" refers to information about the target audience of the advertisement as specified by the advertiser.

[0040] Returning to the explanation of Figure 3, let's describe the statistical information database 32. The statistical information database 32 is a database that stores statistical information. Figure 5 is a diagram showing an example of the statistical information database 32 according to this embodiment.

[0041] As shown in Figure 5, the statistical information database 32 according to this embodiment stores information on items such as "advertising target," "number of times provided," "KPI (Key Performance Indicator)," and "advertising features" in a manner that associates them with each other.

[0042] "Advertised subject" refers to the product or service that is the subject of the advertisement, similar to the "Advertised subject" shown in Figure 4. "Number of times provided" is the number of times the advertisement related to the corresponding advertised subject was provided to the user terminal 10.

[0043] "KPI" is an indicator that shows the performance of the corresponding ad target by the ad (AD), for example, a statistical value of the performance of the ad provided on each user terminal 10. "Ad features" is a statistical value of the features of the ad provided on each user terminal 10.

[0044] Returning to the explanation of Figure 3, let's describe the estimation model database 33. The estimation model database 33 is a database that stores estimation models. The estimation model is a model that estimates KPIs from the content of delivery requests requested by advertisers.

[0045] For example, the estimation model is a simulation model that estimates KPIs based on the content of a delivery request using multiple hypothetically defined personas. Alternatively, the estimation model may be a model that has learned the relationship between the content of the request and the features of the ad provided on the user terminal 10. In this case, the estimation model becomes a simulation model that estimates the ad provided on each user terminal 10 based on the content of the request.

[0046] Next, the control unit 4 will be described. The control unit 4 is, for example, a controller, and is realized by the execution of various programs stored in the storage device inside the information providing device 1 using RAM as the working area, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit), etc.

[0047] As shown in Figure 3, the control unit 4 includes a reception unit 41, an estimation unit 42, a generation unit 43, a provision unit 44, an update unit 45, and a notification unit 46. The reception unit 41 receives various information from each user terminal 10 and advertiser terminal 100.

[0048] The reception unit 41 receives advertising distribution requests from the advertiser terminal 100. For example, the reception unit 41 registers information regarding advertising distribution requests from the advertiser terminal 100 in the advertising information database 31 and passes it to the estimation unit 42.

[0049] Furthermore, the reception unit 41 receives statistical information related to advertising (AD) from each user terminal 10. The reception unit 41 registers the statistical information received from each user terminal 10 into the statistical information database 32.

[0050] The estimation unit 42 performs various estimations regarding the advertisement based on the delivery request from the advertiser terminal 100. For example, the estimation unit 42 inputs information about the delivery request to the estimation model stored in the estimation model database 33, and estimates the KPIs of the advertisement estimated from the content of the delivery request and the ad provided to each user terminal 10. The estimation results from the estimation unit 42 are provided to the advertiser terminal 100 via the provision unit 44.

[0051] The generation unit 43 generates reports for advertisers. For example, the generation unit 43 generates reports for each advertiser based on various statistical information stored in the statistical information database 32.

[0052] For example, the generation unit 43 generates a report for each advertiser that includes information such as the number of times advertising ads were provided, KPIs, and advertising features. The report generated by the generation unit 43 is provided to the corresponding advertiser terminal 100 via the provision unit 44.

[0053] The information provider unit 44 provides various information to each user terminal 10 and advertiser terminal 100. For example, the information provider unit 44 provides advertising information to the user terminal 10. In this case, the information provider unit 44 may provide common advertising information to all user terminals 10, or it may provide different advertising information to each user terminal 10. For example, if the advertised product is alcohol, the information provider unit 44 provides the corresponding advertising information to the user terminal 10 of user U who is over the age of majority. The selection of the advertised product may be made by the user terminal 10.

[0054] Furthermore, the provisioning unit 44 provides the advertiser terminal 100 with the estimation results from the estimation unit 42. These estimation results include information such as the KPIs of the advertisement estimated from the content of the delivery request and information on the distribution of ad feature quantities provided to each user terminal 10.

[0055] This allows advertisers to understand the effectiveness of their ads and their ad placements, and to improve the content of their delivery requests in advance. In this case, advertisers may, for example, specify a delivery area so that the distribution of ad placement features falls within that area. This allows each user terminal 10 to provide ad placements that match the advertiser's image.

[0056] Furthermore, the provisioning unit 44 provides the advertiser terminal 100 with the report generated by the generation unit 43. This allows the advertiser to retrospectively understand the effectiveness of their advertising campaigns.

[0057] [3. Example of user terminal configuration] Next, an example of the configuration of the user terminal 10 will be described using Figure 6. Figure 6 is a block diagram showing an example of the configuration of the user terminal 10 according to the embodiment. As shown in Figure 6, the user terminal 10 according to the embodiment includes a communication unit 110, a storage unit 120, and a control unit 130. The user terminal 10 also has a display unit 111, an operation unit 112, a camera 113, a speaker 114, and a sensor group 115.

[0058] The communication unit 110 is implemented, for example, by a NIC (Network Interface Card). The communication unit 110 transmits and receives information with external devices via various wireless communication networks such as 4G (Generation), 5G, LTE (Long Term Evolution), WiFi (registered trademark), or wireless LAN (Local Area Network), or various wired communication networks.

[0059] The display unit 111 displays images or videos input from the control unit 130. The operation unit 112 receives various operations from the user U. The display unit 111 and the operation unit 112 are implemented, for example, by a touch panel display.

[0060] Camera 113 is a camera installed in the user terminal 10. Camera 113 includes, for example, an out-camera that captures the area outside the screen and an in-camera that captures the area inside the screen.

[0061] The speaker 114 outputs sound input from the control unit 130. The sensor group 115 includes various sensors mounted on the user terminal 10. For example, the sensor group includes an illuminance sensor, a barometric pressure sensor, a temperature sensor, a step counter, a G-sensor, a sound collection sensor, a LiDAR, etc. The user terminal 10 may also, for example, cooperate (pair) with sensors mounted on a wearable device worn by user U to acquire various information such as user U's heart rate, body temperature, stress, and sleep status.

[0062] The storage unit 120 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, or storage devices such as hard disks and optical discs. The storage unit 120 also includes a log information storage unit 121, a camera image storage unit 122, a material information storage unit 123, an advertisement generation model storage unit 124, an advertisement information storage unit 125, and an advertisement evaluation model storage unit 126.

[0063] The log information storage unit 121 stores various log information related to the user U's activity history, etc. Figure 7 shows an example of the log information storage unit 121 according to this embodiment. As shown in Figure 7, the log information storage unit 121 stores information such as "date and time," "log information," "location information," and "environmental information" in association with each other.

[0064] "Date and time" is the date and time when user U performed the action corresponding to the log information. "Log information" is user U's activity history. Here, log information includes, for example, user U's operation history and browsing history of user terminal 10. Alternatively, log information may also be, for example, user U's activity history. User terminal 10 can acquire camera images from camera 113 and estimate user U's activity history by analyzing said camera images. Furthermore, log information may also include payment history of payment applications such as cashless payments (e.g., QR code (registered trademark) payments) using user terminal 10.

[0065] "Location information" refers to information about the place where user U performed the corresponding action. "Environmental information" refers to information about the surrounding environment when user U performed the corresponding action. For example, environmental information includes information such as temperature, humidity, illuminance, and atmospheric pressure, as well as surrounding sounds (music and conversation content), and information about the operating status of home appliances (e.g., air conditioners and other air conditioning equipment). Information about the operating status of home appliances can be obtained, for example, from a smart speaker that is linked to the home appliance.

[0066] Furthermore, the user terminal 10 may be configured to estimate environmental information by taking pictures of its surroundings with the camera 113. For example, the user terminal 10 may estimate the user U's posture, clothing, belongings, surrounding people, surrounding objects, etc., and include them in the environmental information.

[0067] Returning to the explanation of Figure 6, the camera image storage unit 122 will be described. The camera image storage unit 122 stores camera images captured by the camera 113. Figure 8 is a diagram showing an example of the camera image storage unit 122 according to this embodiment.

[0068] As shown in Figure 8, the camera image storage unit 122 stores information such as "image ID," "shooting date and time," "shooting location," "image information," "subject information," and "shooting environment" in a manner that associates them with each other.

[0069] The "Image ID" is an identifier used to identify each camera image. The "Date and Time of Capture" and "Location of Capture" are the date and time and location of capture for the camera image identified by the corresponding Camera ID.

[0070] "Image information" refers to the image information of the camera image identified by the corresponding camera ID. "Subject information" refers to information about the subject captured in the camera image identified by the corresponding camera ID.

[0071] "Shooting environment" refers to information about the environment in which the camera image, identified by the corresponding camera ID, was taken. For example, "environmental information" refers to information about the surrounding environment when the camera image was taken, and includes information such as temperature, humidity, illuminance, atmospheric pressure, and surrounding sounds (music, conversations, etc.).

[0072] Returning to the explanation of Figure 6, let's describe the material information storage unit 123. The material information storage unit 123 stores material information. For example, material information is information that is used as material when generating an advertisement AD on the user terminal 10, and includes product images for each advertisement, the advertiser's logo image and character image, images of celebrities used in the advertisement, and music used in the advertisement.

[0073] The ad generation model storage unit 124 stores the ad generation model. The ad generation model is a model that has been trained to generate ads for a specified target using user information held by the user terminal 10.

[0074] The advertising information storage unit 125 stores advertising information. The advertising information includes information about advertisements generated by the advertising generation model and information about the evaluation of the corresponding advertisements. Figure 9 shows an example of the advertising information storage unit 125 according to this embodiment.

[0075] As shown in Figure 9, the advertising information storage unit 125 stores information on items such as "advertising ID," "advertising target," "advertising type," "advertising information," "features," "provision conditions," and "evaluation information" in a manner that associates them with each other.

[0076] The "Ad ID" is an identifier used to identify each ad. The "Target" is the target audience for the ad identified by the corresponding Ad ID. The "Ad Type" is the type of ad identified by the corresponding Ad ID. For example, "Ad Type" includes banner ads, text ads, search ads, etc.

[0077] "Advertising information" refers to the advertising information (data body) of an ad identified by its corresponding advertising ID. "Features" are the features of an ad identified by its corresponding advertising ID, and may include, for example, a vector corresponding to the features or text describing the characteristics of the ad.

[0078] "Provision conditions" are the conditions under which the ad AD identified by the corresponding ad ID is provided to user U. "Evaluation information" is evaluation information for the ad AD identified by the corresponding ad ID. For example, various metrics such as KPIs may be used as evaluation information.

[0079] Returning to the explanation of Figure 6, let's describe the control unit 130. The control unit 130 is, for example, a controller, and is realized by the execution of various programs stored in the storage device inside the user terminal 10 using RAM as the working area, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit), etc.

[0080] As shown in Figure 6, the control unit 130 includes a reception unit 131, an analysis unit 132, a selection unit 133, a generation unit 134, an evaluation unit 135, and a supply unit 136.

[0081] The reception unit 131 receives advertising information from the information provision device 1 via the communication unit 110. The advertising information includes, for example, information about the materials used in the advertisement, the image of the advertisement, and the generation conditions when generating the advertisement.

[0082] Furthermore, the reception unit 131 receives target information indicating the target of the advertisement, and trend information indicating the tendency of the image to be combined with the target information.

[0083] The analysis unit 132 analyzes the camera image captured by the camera 113. For example, the analysis unit 132 analyzes the subject in the camera image by performing image analysis of the camera image. For example, the subject may include people, facial expressions, clothing, belongings, and surrounding circumstances in the camera image.

[0084] Furthermore, the analysis unit 132 analyzes the surrounding environment of the user terminal 10 based on various information input from the sensor group 115. For example, the surrounding environment includes temperature, illuminance, atmospheric pressure, humidity, and ambient noise.

[0085] Furthermore, the analysis unit 132 analyzes various data stored in the user terminal 10 and generates log information regarding the user U's operation history and behavior history of the user terminal 10. Alternatively, the user terminal 10 may capture background images of its surroundings using the camera 113, and the analysis unit 132 may analyze the camera images to determine the user U's behavior history.

[0086] Furthermore, the analysis unit 132 may, for example, take a background image of the display screen of the display unit 111 and analyze the message sending and receiving history. Alternatively, the analysis unit 132 may analyze the mental state of user U based on various information input from the wearable device worn by user U.

[0087] The selection unit 133 selects an advertising target from the material information storage unit 123 according to the current situation of user U. For example, the selection unit 133 inputs information about user U's situation and surrounding circumstances, which is analyzed by the analysis unit 132, into the advertising evaluation model to select the advertising target and material information that user U desires in the current situation. In other words, the selection unit 133 uses the advertising evaluation model, which is a person model of user U, to select recommended products, etc., as advertising targets.

[0088] This allows the selection unit 133 to select advertising targets that are of high value to user U. The selection unit 133 may also select a combination of multiple advertising targets. For example, the selection unit 133 may select a combination of multiple advertising targets, such as a car and a BBQ set.

[0089] In this case, the user terminal 10 will generate an ad (AD) including a car and a BBQ set and provide it to user U. In other words, the user terminal 10 will be able to provide an ad (AD) that is expected to have a synergistic effect depending on the combination of advertised items.

[0090] Furthermore, the selection unit 133 selects an image from the information held in the user terminal 10 that corresponds to the trend information received by the reception unit 131.

[0091] For example, the selection unit 133 selects a camera image from the camera image storage unit 122 that corresponds to the context included in the trend information. For example, if the context included in the trend information is "memories with my father," the selection unit 133 selects a camera image from the camera image storage unit 122 that corresponds to "memories with my father."

[0092] Furthermore, the selection unit 133 may select camera images corresponding to various conditions specified by trend information. These conditions include shooting location, shooting environment, shooting time, and conditions related to the subject.

[0093] Conditions regarding the subject include specifying the people and attributes in the camera image, such as a camera image showing user U and user U's father, or a camera image showing only user U's father; specifying user U's facial expression, clothing, and belongings; and specifying scenery, such as a sunset.

[0094] Furthermore, the selection unit 133 may select an image corresponding to trend information based on text information held by the user terminal 10. For example, the selection unit 133 may select a screen image related to the message sending and receiving history using various messaging applications as the image.

[0095] If the context included in the trend information is "memories with my father," the selection unit 133 opens the messaging app in the background, generates a screen image of the message sending and receiving history that corresponds to "memories with my father" from the message sending and receiving history between the user and their father, and selects the image corresponding to the trend information from the screen image.

[0096] Furthermore, the selection unit 133 may, for example, select images to be used in the advertisement based on the subject of the advertisement. For example, if the subject of the advertisement is "beer," the selection unit 133 selects images to be used in the advertisement, such as camera images in which "beer" is visible, or camera images that evoke the image of "beer" for the user U.

[0097] The generation unit 134 generates an ad for a specified target using user information held by the user terminal 10 and an ad generation model that has been trained to generate ads for the specified target.

[0098] For example, the generation unit 134 generates an ad by inputting the material information of the ad target selected by the selection unit 133 and the user information held by the user terminal 10 into the ad generation model. For example, the user information here is a generation parameter generated by the ad evaluation model based on the user information, and the generation parameter is a parameter set in the ad generation model so that an AI image with high appeal to user U is generated.

[0099] The ad generated by the ad generation model includes AI image G, text information T, and stock image P (see Figure 1). For example, as mentioned above, the generation parameters are updated as needed, so the ad generation model can optimize the AI ​​image G, text information T, and stock image P individually.

[0100] In this case, the generation unit 134 may, for example, generate an ad by further inputting the context included in the trend information specified by the advertiser into the ad generation model. In this case, the generation unit 134 can generate the ad requested by the advertiser while taking into account the preferences of user U, etc.

[0101] Furthermore, the generation unit 134 generates an advertisement (AD) using the image selected by the selection unit 133. Figure 10 is a schematic diagram showing an example of advertisement generation using a camera image according to the embodiment.

[0102] As shown in Figure 10, the selection unit 133 selects a camera image corresponding to the trend information from the camera image storage unit 122 and a source image from the source information storage unit 123. The selection unit 133 may also select multiple camera images or multiple source images.

[0103] Furthermore, when generating a video-format advertising ad, the selection unit 133 may select a video from the camera image storage unit 122, or it may select multiple camera images according to the length of the video.

[0104] Next, the generation unit 134 generates an ad by inputting the camera image, source image, and user information (e.g., generation parameters) into the ad generation model. In this case, the generation unit 134 may use the camera image as is for the ad, or it may process the camera image before using it for the ad.

[0105] Furthermore, for example, the generation unit 134 may generate and provide advertising ads according to the surrounding environment. Figure 11 is a schematic diagram showing an example of providing advertising ads according to the surrounding environment according to the embodiment.

[0106] For example, the analysis unit 132 analyzes the surrounding environment based on the background images captured by the camera 113 and the detection results of the sensor group 115, and the generation unit 134 generates an ad by further inputting parameters related to the analysis results of the surrounding environment into the ad generation model.

[0107] In this case, advertising ads (ADs) are generated according to the surrounding environment and provided to the user U according to the surrounding environment. For example, the generation unit 134 generates advertising ads that match the surrounding music and the surrounding space.

[0108] Furthermore, the generation unit 134 may generate advertising ads (ADs) that correspond to the music or video content being played by the user terminal 10. In this case, the user terminal 10 can provide user U with advertising ads that match the content that user U is watching.

[0109] Returning to the explanation of Figure 6, let's describe the evaluation unit 135. The evaluation unit 135 evaluates the advertisements generated by the generation unit 134 using an advertisement evaluation model that has been trained to estimate the evaluation of advertisements by user U on the user terminal 10.

[0110] The evaluation unit 135 estimates user U's evaluation of an advertisement by inputting the information about the advertisement (AD) generated by the generation unit 134 into the advertisement evaluation model. For example, the advertisement evaluation model is a person model of user U that has been trained using various log information of user U, and is a model that has been trained to estimate user U's evaluation of an advertisement when an advertisement (AD) is input. The advertisement evaluation model is updated each time new log information is added to the log information storage unit 121.

[0111] The evaluation unit 135 evaluates each ad (AD) before it is provided at predetermined intervals and passes the evaluation results to the provision unit 136 each time. The evaluation unit 135 also evaluates the performance of the ad based on log information after the ad has been provided.

[0112] For example, the performance of advertising (AD) is evaluated based on various indicators such as KPIs. The evaluation unit 135 then updates the generation parameters of the ad generation model based on the measurement results of the advertising (AD) performance and places an order for advertising (AD) with the generation unit 134.

[0113] The supply unit 136 provides the advertising ad generated by the generation unit 134 when the evaluation result by the evaluation unit 135 meets predetermined conditions. For example, the supply unit 136 provides the corresponding advertising ad when the score, which is the evaluation result by the evaluation unit 135 of the advertising ad generated by the generation unit 134, exceeds a predetermined value.

[0114] This allows the delivery unit 136 to provide advertising ads (ADs) to user U at a time when they are most likely to be effective. In this way, the user terminal 10 optimizes the target of the advertisement, the advertising ads, and the timing of their delivery using various information about user U that can be obtained on the user terminal 10.

[0115] On the user terminal 10, it becomes possible to optimize advertising (AD) using sensitive information that is not stored on the server in conventional advertising distribution systems.

[0116] Furthermore, the provision unit 136 provides statistical information to the information provision device 1. For example, the statistical information consists of minimal information such as information on the features of the advertisement AD provided to user U and the results of performance measurement, and does not include sensitive information that would identify user U personally.

[0117] In other words, the provision unit 136 can protect user U's personal information by providing only the minimum necessary information regarding the advertisement AD provided to user U as statistical information to the information provision device 1.

[0118] [4. Processing Flow] Next, the processing procedure performed by the information providing device 1 according to the embodiment will be described using Figure 12. Figure 12 is a flowchart showing an example of the processing procedure performed by the information providing device 1 according to the embodiment.

[0119] As shown in Figure 12, first, the information provider 1 receives a request for advertising distribution from the advertiser terminal 100 (step S101). Next, the information provider 1 provides advertising information to each user terminal 10 (step S102).

[0120] Next, the information providing device 1 collects statistical information about advertising (AD) from the user terminal 10 (step S103), and generates reports for each advertiser based on the collected statistical information (step S104).

[0121] Then, the information provider 1 provides the generated report for each advertiser to the corresponding advertiser terminal 100 (step S105), and terminates the process.

[0122] Next, Figures 13 to 16 will be used to explain the processing procedure performed by the user terminal 10. Figures 13 to 16 are flowcharts showing an example of the processing procedure performed by the user terminal 10 according to this embodiment.

[0123] First, we will explain the series of processes related to ad delivery using Figure 13. As shown in Figure 13, the user terminal 10 acquires various log information about user U (step S201).

[0124] Next, the user terminal 10 selects an advertising target based on the analysis results of the log information (step S202) and determines the generation parameters of the advertising generation model (step S203). Subsequently, the user terminal 10 generates an advertising ad based on the material information of the advertising target selected in step S202 and the generation parameters determined in step S203 (step S204).

[0125] Next, the user terminal 10 evaluates the ad generated in step S204 (step S205), provides the ad at a timing when the score resulting from the evaluation exceeds a threshold (step S206), and then terminates the process.

[0126] Next, Figure 14 will be used to explain the series of processes after the provision of the advertising ad. As shown in Figure 14, after the provision of the advertising ad, the user terminal 10 acquires log information after the provision of the advertising ad (step S211).

[0127] Next, the user terminal 10 evaluates the ad provided to user U based on the acquired log information (step S212). Subsequently, the user terminal 10 updates the generation parameters of the ad generation model based on the evaluation result of the ad (step S213), and terminates the process.

[0128] Next, Figure 15 will be used to explain the processing procedure for generating advertisements that include AI images. As shown in Figure 15, the user terminal 10 selects the advertisement target (step S221) and determines the material information (step S222).

[0129] Then, the user terminal 10 inputs the material information and generation parameters into the ad generation model to generate an ad AD including the AI ​​image G (step S223), and the process ends.

[0130] Next, the processing procedure for generating advertisements including camera images will be explained using Figure 16. As shown in Figure 16, the user terminal 10 selects the advertisement target (step S231) and determines the material information (step S232).

[0131] Next, the user terminal 10 selects a camera image corresponding to the trend information (step S233). Then, the user terminal 10 generates an advertisement AD including the camera image (step S234) and terminates the process.

[0132] [5. Variations] The above-described embodiment explains the case in which the user terminal 10 generates advertisements (ADs) to be provided to user U, but it is not limited to this. For example, the user terminal 10 may generate various types of content, such as videos, using the method of the present invention. In other words, the purpose of providing content is not limited to advertisements, and the user terminal 10 may generate and evaluate content provided for any purpose.

[0133] [6. Effects] The information processing program according to the above embodiment causes the computer in the terminal device to execute a generation procedure that generates an advertisement for a specified target using user information held by the terminal device and an advertisement generation model that has been trained to generate an advertisement for a specified target; an evaluation procedure that evaluates the advertisement generated by the generation procedure using an advertisement evaluation model that has been trained to estimate the evaluation of the advertisement by the terminal device's user; and a provision procedure that provides the advertisement generated by the generation procedure if the evaluation result from the evaluation procedure satisfies predetermined conditions.

[0134] Furthermore, the generation procedure generates the advertisement using advertising materials specified by the advertiser. Furthermore, the evaluation procedure evaluates the advertisement using an advertising evaluation model, which is a user person model estimated from user information.

[0135] Furthermore, the evaluation procedure evaluates advertisements using an advertising evaluation model, which is a user model estimated from the terminal device's operation history.

[0136] Furthermore, the generation procedure evaluates advertisements using an ad evaluation model, which is a user person model estimated from the video or music playback history of the terminal device. The evaluation procedure further evaluates the advertisements after they have been delivered and updates the generation parameters related to ad generation by the generation procedure based on the evaluation results.

[0137] Furthermore, the evaluation procedure updates the evaluation results of the advertisement generated by the generation procedure according to the surrounding environment of the terminal device, and the provision procedure provides the advertisement generated by the generation procedure if the updated evaluation results from the evaluation procedure meet predetermined conditions.

[0138] Furthermore, the generation procedure generates advertisements that are appropriate to the surrounding environment of the terminal device. The generation procedure generates advertisements related to the advertising targets selected based on the evaluation results from the evaluation procedure. Furthermore, the generation procedure generates advertisements using image information held by the terminal device. Furthermore, the provision procedure provides the characteristic information of the advertisements provided to the user to an external device.

[0139] Through any or a combination of the above-described processes, the information processing program relating to this application can deliver appropriate advertisements to users while protecting personal information.

[0140] [7. Hardware Configuration] Furthermore, the information providing device 1 and user terminal 10 according to the above embodiment are realized by a computer 1000 having a configuration such as that shown in Figure 17. Figure 17 is a hardware configuration diagram showing an example of a computer that realizes the functions of the user terminal 10 according to the embodiment. The computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0141] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, controlling various components. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0142] The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 receives data from other devices via the network (communication network) N and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the network N.

[0143] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice (in Figure 17, output devices and input devices are collectively referred to as "input / output devices") via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the generated data to output devices via the input / output interface 1600.

[0144] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it 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) or 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.

[0145] For example, when computer 1000 functions as a user terminal 10 according to the embodiment, the CPU 1100 of computer 1000 realizes the functions of the control unit 130 by executing programs loaded on RAM 1200. The CPU 1100 of computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via the network N.

[0146] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0147] [8. Other] Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0148] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0149] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0150] Furthermore, the terms "section, module, unit" mentioned above can be replaced with "means" or "circuit." For example, the acquisition unit can be replaced with acquisition means or acquisition circuit. [Explanation of Symbols]

[0151] 1 Information provision device 10. User terminals 31 Advertising Information Database 32 Statistical Information Databases 33 Estimated Model Database 41 Reception Department 42 Estimation part 43 Generation part 44 Providing Department 45 Update section 46 Notification Department 100 advertiser terminals 110 Communications Department 111 Display section 112 Operation section 113 Camera 114 speakers 115 Sensor Groups 121 Log Information Storage Unit 122 Camera image storage unit 123 Material Information Storage Unit 124 Ad generation model memory unit 125 Advertising Information Storage Unit 126 Advertising Evaluation Model Memory Unit 130 Control Unit 131 Reception Department 132 Analysis Department 133 Selection Section 134 Generation part 135 Evaluation Department 136 Provision Department

Claims

1. The computer in the terminal device A procedure for obtaining target information that identifies the items to be provided to the user from an external device, A generation procedure for generating content to be provided to a user who owns the terminal device, using a content generation model that has been trained to generate content relating to user information held by the terminal device and the target recipient identified by the target information, An evaluation procedure for evaluating the content generated by the generation procedure on the terminal device using a content evaluation model that has been trained to estimate the evaluation of content by users of the terminal device, If the evaluation results from the evaluation procedure satisfy predetermined conditions, a provision procedure is provided to the user for the content generated by the generation procedure. Make it run, The aforementioned provision procedure is, To provide the feature quantities of the content to the external device without transmitting the user information to the external device. An information processing program characterized by the following.

2. The acquisition procedure described above is: The material information or trend information of the content is obtained from the external device as the target information. The aforementioned generation procedure is: To generate the content based on the aforementioned material information or the aforementioned trend information. The information processing program according to claim 1, characterized by the following:

3. The aforementioned evaluation procedure is: The content is evaluated using the content evaluation model, which has been learned based on the operation history, image information, location information, or sensor output of the terminal device. The information processing program according to claim 1, characterized by the following:

4. The aforementioned generation procedure is: To generate the content using the image information held by the terminal device. The information processing program according to claim 1, characterized by the following:

5. The aforementioned generation procedure is: The content is generated according to the surrounding environment of the terminal device, obtained by analyzing the output results of the sensors that the terminal device has. The information processing program according to claim 1, characterized by the following:

6. The aforementioned evaluation procedure is: Evaluate the content based on the user's behavior history with respect to the content after it has been provided, and update the generation parameters in the generation procedure based on the evaluation results. The information processing program according to claim 1, characterized by the following:

7. An acquisition unit that acquires target information from an external device to identify the items to be provided to the user, A generation unit generates the content to be provided to a user who owns the terminal device, using a content generation model that has been trained to generate content relating to user information held by the terminal device and the target recipient identified by the target information, An evaluation unit evaluates the content generated by the generation unit on the terminal device using a content evaluation model that has been trained to estimate the evaluation of content by users of the terminal device. When the evaluation result by the evaluation unit satisfies predetermined conditions, the provision unit provides the content generated by the generation unit to the user. Equipped with, The aforementioned supply unit is, To provide the feature quantities of the content to the external device without transmitting the user information to the external device. A terminal device characterized by the following.

8. A method of information processing performed by a computer in a terminal device, The acquisition process involves obtaining target information from an external device to identify the items to be provided to the user, and A generation step in which the terminal device generates the content to be provided to a user who owns the terminal device, using a content generation model that has been trained to generate content relating to user information held by the terminal device and the target recipient identified by the target information, An evaluation step in which the content generated in the generation step is evaluated on the terminal device using a content evaluation model that has been trained to estimate the evaluation of content by users of the terminal device, If the evaluation results from the evaluation step satisfy predetermined conditions, the provision step provides the content generated in the generation step to the user. Have the computer run it, The aforementioned provisioning process is, To provide the feature quantities of the content to the external device without transmitting the user information to the external device. An information processing method characterized by the following.

Images