Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device, an information processing method, and an information processing program, capable of reducing the work load required to estimate a user's impression of content.SOLUTION: An information processing device comprises a reception unit, a generation unit, and a provision unit. The reception unit receives: a selection of a target content which is an estimate target for impressions that users would have or the users have come to have; and a selection of an impression group which is a group of the impressions. The generation unit generates, by using a generative AI, the impression content indicative of the impressions which are estimated that the users would have or the users have come to have with respect to the target content among the plurality of impressions included in the impression group received by the reception unit. The provision unit provides the impression content generated by the generation unit.SELECTED DRAWING: Figure 3

Description

Technical Field

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

Background Art

[0002] Conventionally, content creators such as web content try to create content so that the impression that users have of the content matches the target impression. However, there is a possibility that the impression that users actually have of the content may deviate.

[0003] Patent Document 1 discloses a technique in which an evaluator is made to input an evaluation level of an impression received from viewing content, and a pair of the content and the evaluation level of the impression received by the evaluator for this content is used as teaching data for learning a machine learning model.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the above prior art, since the evaluator needs to input the evaluation level of the impression, there is a problem that the work load is high.

[0006] The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program capable of reducing the work load required to estimate the impression of a user on content.

Means for Solving the Problems

[0007] The information processing apparatus according to the present application generates a reception unit, a generation unit, and a provision unit. The reception unit receives the selection of target content to be estimated for the impression held or had by the user and the selection of an impression group which is a group of impressions. The generation unit generates impression content indicating an impression that the user is estimated to hold or have had with respect to the target content among the plurality of impressions included in the impression group received by the reception unit, using a generation AI. The provision unit provides the impression content generated by the generation unit.

Effect of the Invention

[0008] According to one aspect of the embodiment, there is an effect that the workload required to estimate the user's impression of the content can be reduced.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Best Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments for implementing the information processing apparatus, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing apparatus, information processing method, and information processing program according to the present application are not limited by this embodiment. Also, the respective embodiments can be appropriately combined within a range that does not conflict with the processing contents. Further, in the following respective embodiments, the same parts are denoted by the same reference numerals, and redundant explanations are omitted.

[0011] 〔1. An Example of Information Processing〕 First, an example of information processing according to the embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram for explaining the information processing according to the embodiment.

[0012] The information processing apparatus 1 shown in FIG. 1 is an information processing apparatus that performs, for example, generation of impression content that is content indicating an impression that a user U has or is presumed to have had with respect to target content that is the target content. It is realized by, for example, one or more servers or a cloud system, etc. The user U is a user of the terminal device 2 and a user of an online service to which the target content is provided.

[0013] As shown in FIG. 1, the information processing apparatus 1 receives an impression content generation request transmitted from the terminal device 3 of an operator O who creates or provides the target content (step S1). The impression content generation request includes the target content that is the target of estimation of the impression that the user U has or has had, information indicating an impression group that is a group of impressions, information specifying the estimation method, information indicating the target content type, and information indicating the impression content type.

[0014] The target content, impression group, estimation method, target content type, and impression content type specified in the impression content generation request are, for example, the target content, impression group, estimation method, target content type, and impression content type selected by the operator O.

[0015] By receiving an impression content generation request, the information processing apparatus 1 receives selections of the target content, impression group, estimation method, target content type, and impression content type. Note that the operator O is an employee of a company that creates or provides the target content, but may also be an employee of a company commissioned by such a company.

[0016] The target content is web content, and examples include, but are not limited to, web pages, catchcopy, creatives, and stamps. The target content is, for example, content that is a candidate for or target of provision on an EC (Electronic Commerce) site, news site, restaurant introduction site, image posting site, video viewing site, etc., and includes advertising content, but is not limited to such examples. The EC site is, for example, a shopping site, an auction site, etc., but is not limited to such examples.

[0017] The target content is content that can be provided on an online site provided by the information processing apparatus 4, and the user U can operate the terminal device 2 to use the online site provided by the information processing apparatus 4. The online site provided by the information processing apparatus 4 is an EC site, news site, store introduction site, image posting site, video viewing site, etc., but is not limited to such examples.

[0018] The information processing apparatus 4 receives, as submission content, product content such as product introduction texts indicating the content of product introductions from the operator O etc. on the EC site, and provides such submission content to each user U. Note that the product introduction text can also be referred to as a product description text. Also, the submission content on the EC site includes, in addition to the product content, comments and evaluations etc. posted for such products.

[0019] Also, the information processing apparatus 4 receives, as submission content, news articles from the operator O etc. on the news site, and provides such submission content to each user U. On the news site, the submission content on the news site includes, in addition to the news articles, comments etc. posted for such news articles.

[0020] Also, the information processing apparatus 4 receives, as submission content, store content such as store introduction texts indicating the content of store introductions from the operator O on the store introduction site, and provides such submission content to each user U. Also, the submission content on the store introduction site includes, in addition to the store content, reviews and comments etc. posted for such stores. The store introduction site is, for example, a site for each store type, such as a restaurant introduction site.

[0021] Also, the information processing apparatus 4 receives, as submission content, image content (for example, still image content or video content) etc. from the operator O etc. on the image submission site, and provides such submission content to each user U. Also, the information processing apparatus 4 receives, as submission content, video content and title and caption text etc. in the video content on the video viewing site, and provides such submission content to each user U.

[0022] The impression groups specified by the impression content generation request are impression groups designated by the user U from among a plurality of impression groups C1 to Cn (n is an integer of 2 or more). The plurality of impression groups C1 to Cn are, for example, impression groups classified according to the values of the user U, impression groups classified according to the preferences of the user U, impression groups classified according to the lifestyle of the user U, etc., but are not limited to such examples.

[0023] In the example shown in FIG. 1, the impression groups according to the values of the user U are impression groups that classify the impressions held or had by the user U as impressions based on Schwartz's value theory. They are classified according to 10 classification conditions: power, achievement, hedonism, stimulation, self-determination, universalism, benevolence, tradition, harmony, and security. In the example shown in FIG. 1, for the impression groups according to the values based on Schwartz's value theory, each of the 10 classification conditions is further subdivided into a plurality of classification conditions, and the impressions held or had by the user U are classified according to 56 classification conditions.

[0024] Specifically, the classification condition "power" is subdivided into 5 classification conditions: "social approval", "wealth", "authority", "social power", and "maintenance of face". The classification condition "achievement" is subdivided into 5 classification conditions: "intelligence", "competence", "success", "ambition", and "influence". The classification condition "hedonism" is subdivided into 2 classification conditions: "enjoyment of life" and "fun". The classification condition "stimulation" is subdivided into 3 classification conditions: "courage", "life with change", and "lively life".

[0025] Also, the classification condition "self-determination" is subdivided into 6 classification conditions: "freedom", "independence", "curiosity", "creativity", "self-choice", and "self-respect". The classification condition "universalism" is subdivided into 9 classification conditions: "wisdom", "world peace", "beautiful world", "social justice", "inner harmony", "environmental protection", "equality", "generosity", and "harmony with nature". The classification condition "benevolence" is subdivided into 9 classification conditions: "true friendship", "meaning of life", "responsibility", "loyalty", "mature love", "assistance", "honesty", "tolerance", and "spiritual world".

[0026] In addition, the classification condition "harmony" is subdivided into four classification conditions: "correctness of manners", "self-training", "respect for parents and elders", and "obedience". The classification condition "tradition" is subdivided into six classification conditions: "humility", "detachment", "respect for tradition", "piety", "steadiness", and "acceptance of fate". The classification condition "safety" is subdivided into seven classification conditions: "health", "safety of family", "social order", "cleanliness", "return of kindness", "sense of belonging", and "national security".

[0027] The values of the user U are not limited to the examples described above. For example, they may be values classified by traditionalism, success orientation, self-actualization orientation, symbiosis orientation, etc., or other values. For example, when the object is a car, the values of the user U may be values classified by, for example, emphasis on safety, emphasis on habitability, emphasis on design, emphasis on drivability, etc. When the object is a house, the values of the user U may be values classified by, for example, emphasis on location, emphasis on price, emphasis on floor plan, emphasis on design, emphasis on future potential, etc.

[0028] The impression groups according to the lifestyle of the user U are, for example, impression groups by the AIO (Activities Interest Opinions) approach or impression groups by VALS (Values And LifeStyles). The impression group by the AIO approach is an impression group classified by one or more impression groups among activities, interests, opinions, etc. The impression group by VALS is, for example, an impression group classified by self-actualizers, achievers, group belongers, life sustainers, life strugglers, success aspirants, socially conscious people, intellectuals, and young intellectuals.

[0029] In addition, the impression group may be, for example, an impression group shown by the innovator theory that classifies the process of the spread of new products. This impression group is an impression group classified by innovators, early adopters, early majorities, late majorities, and laggards.

[0030] There are a plurality of estimation methods including a first estimation method and a second estimation method for estimating the impression that user U has or had about the target content. The first estimation method is a method of directly estimating by the generation AI (Artificial Intelligence) the impression that user U has or had about the target content. The second estimation method is a method of causing the generation AI to generate reference content that is estimated to be held by user U for each impression included in the impression group, and comparing such reference content with the target content to estimate the impression that user U has or had about the target content.

[0031] The target content type is, for example, the type of the target content, and is indicated by, for example, a combination of an online site and a content type. For example, the target content type includes, but is not limited to, product descriptions on product pages of EC sites, news articles on news sites, comments on news articles on news sites, image content on image posting sites, titles and caption texts of video content on video viewing sites, etc.

[0032] The impression content type is the type of impression content, and includes, for example, a first content type and a second content type. The first content type is the type of impression content that includes information indicating the impression estimated to be held by user U about the target content, and is the type of impression content that supports the creation and consideration of the target content. The second content type is the type of impression content that includes information indicating the impression estimated to be held by user U about the target content, and is the type of impression content that supports the understanding of the impression that user U has about the target content.

[0033] Subsequently, the information processing apparatus 1 estimates, using the generation AI, the impression that is estimated to be held or had by user U about the target content received in step S1 among the plurality of impressions included in the selected impression group which is the impression group received in step S1 (step S2).

[0034] The generative AI used in step S2 is, for example, text generation AI, image generation AI, multimodal AI, etc. Such generative AI is, for example, arranged in an external information processing device, and the information processing device 1 causes the generative AI to generate reference content for each impression via an API (Application Programming Interface) provided by the external information processing device and obtains it, but it is not limited to such an example. For example, the generative AI may be arranged within the information processing device 1.

[0035] The text generation AI is, for example, a language model trained to estimate and output the next token from the input token sequence, and is, for example, a transformer-based model or an RNN (Recurrent Neural Network)-based model.

[0036] The transformer-based model is, for example, GPT (Generative Pre-trained Transformer), BARD (Bidirectional Auto Regressive Dialogues), etc., but is not limited to such examples. The RNN-based model is, for example, RWKV (Receptance Weighted Key Value), etc., but is not limited to such examples. Note that it is desirable to perform learning so that the input information is not used as a new answer to conceal information such as the input personal information.

[0037] The image generation AI is, for example, StackGAN (Generative Adversarial Networks), AttnGAN, T2I (Text-to-Image) with Transformers, DALL-E, etc., but is not limited to such examples. The multimodal AI is, for example, a model that generates an image from text or generates text from an image, and is, for example, GPT-4V, CM3Leon (Chameleon Multimodal Model), etc., but is not limited to such examples.

[0038] In step S2, the information processing apparatus 1 estimates, using a generation AI, an impression or impressions that the user U has of the target content from among a plurality of impressions included in the selected impression group. This will be specifically described below. In the following, it is assumed that the plurality of impressions included in the selected impression group are a plurality of impressions IM1 to IMm (m is an integer of 2 or more). Also, when each of the plurality of impressions IM1 to IMm is shown without being individually distinguished, it may be described as impression IM.

[0039] First, the case where the estimation method and impression content type selected by the information processing apparatus 1 in step S1 are the first estimation method and the first content type will be described. In this case, the information processing apparatus 1 causes the generation AI to estimate an impression IM that the user U has of the target content from among a plurality of impressions IM1 to IMm included in the selected impression group.

[0040] For example, the information processing apparatus 1 inputs, as input information to the generation AI, instruction information instructing to select an impression IM that the user U has of the target content from among a plurality of impressions IM1 to IMm indicated by impression information, impression information indicating a plurality of impressions IM1 to IMm included in the selected impression group, and the target content, and causes the generation AI to output information indicating one or more impressions IM estimated to be had by the user U of the target content.

[0041] For example, assume that the selected impression group is an impression group based on the values of the user U, and the target content has a catchphrase "Because it is used every day, the difference can be noticed." In this case, the information processing apparatus 1 includes, for example, information such as the character string "Please select three or more images that fit the catchphrase 'Because it is used every day, the difference can be noticed.' from the following values. The output format should be in csv format as follows.\nValue sense, Reason" as instruction information in the input information.

[0042] In addition, the information processing device 1 includes, as input information, information including information of the character string "maintenance of social power, authority, wealth, face, social approval, success, ability, ambition, influence, intelligence, enjoyment, enjoyment of life, courage, life with change, lively life, creativity, curiosity, freedom, goal selection, self-respect, independence, environmental protection, beautiful world, harmony with nature, generosity, social justice, wisdom, equality, world peace, inner harmony, assistance, honesty, tolerance, loyalty, responsibility, true friendship, spiritual world, mature love, meaning of life, propriety, respect for parents and elders, self-discipline, obedience, piety, acceptance of fate, humility, moderation, respect for tradition, transcendence, cleanliness, national security, social order, family security, gratitude, health, sense of belonging" as impression information.

[0043] In addition, the information processing device 1 can use, for example, information of the character string "Please select three or more images that fit the following catchphrase from the following values." replaced with information of the character string "Please select three or more impressions that the user is presumed to have for the following catchphrase from the following values." or information of the character string "Please estimate the score indicating the degree of impression that the user is presumed to have for the following catchphrase for each of the following values. The score should be in the range of 1 to 10, and the higher the degree of impression, the larger the value.", but is not limited to such examples. Also, the output format is not limited to the csv format.

[0044] In addition, the instruction information includes character information corresponding to the target content type selected in step S1. In the above example, since the target content type is a catchphrase, the information of the character string "catchphrase" is included in the instruction information. However, when the target content type is a product description on an EC site product page, the instruction information includes the information of the character string "product description".

[0045] In this way, the information processing device 1 generates instruction information corresponding to the target content type by including character information corresponding to the target content type in the instruction information, but it is also possible to input instruction information that does not depend on the target content type to the generation AI.

[0046] For example, the information processing device 1 can set, as a system message, instruction information including the information of the character string "Please select three or more images that fit the following catchphrase from among the following values instead of the character string 'Because it is something used every day, the difference can be noticed.'". The system message is a message that instructs the behavior of the generative AI. For example, in the GPT model of OpenAI, it is a message set in the input information as the content of "role": "system".

[0047] In addition, the information processing device 1 can also estimate, using the generative AI, the impressions held by the user U for the target content by limiting the attributes of the user U, etc., from among a plurality of impressions IM1 to IMm included in the selected impression group. In this case, the information processing device 1 can use, for example, information such as "Please select three or more impressions that are estimated to be held by a 20-year-old male for the following catchphrase from among the following values instead of the information of the character string 'Please select three or more images that fit the following catchphrase from among the following values.'" or the information of the character string "Please estimate the score indicating the degree of the impression that is estimated to be held by a 20-year-old male for the following catchphrase for each of the following values. The score ranges from 1 to 10, and the higher the degree of the impression, the larger the value." When estimating, using the generative AI, one or more impressions IM held by a user U with an attribute other than a 20-year-old male, it is possible to replace the information of the character string "20-year-old male" with information indicating another attribute.

[0048] In addition, when the target content is image content, the information processing device 1 can input information including the target content and instruction information such as the information of the character string "Please select three or more images that fit the input image from the following values." as input information into the multimodal AI, and output the similarity from the multimodal AI. Note that the information processing device 1 can limit the attributes of the user U and output a score indicating the degree of the impression IM in the same manner as when the target content is other than image content.

[0049] Next, a case where the estimation method and the impression content type selected by the information processing device 1 in step S1 are the first estimation method and the second content type will be described. In this case, the information processing device 1 collects action content, which is content indicating the behavior of the user U related to the target content, from the information processing device 4 or the like.

[0050] The action content is content generated by the action of the user U related to the target content. For example, it is posted content such as reviews, comments, and messages posted by the user U regarding the target content. For example, when the target content is a product description or an image of a product, the posted content as the action content is a review or comment posted on the product introduced by the target content.

[0051] Also, when the target content is a store description or an image in the store, the posted content as the action content is a review or comment posted on the store introduced by the target content. Also, when the target content is video content, the posted content as the action content is a review or comment posted regarding the video content.

[0052] In addition, the posted content as action content may be content such as conversations and posts by the user U on the official accounts of products and stores indicated by the target content in the communication application. Also, the posted content as action content may be, for example, content such as posts and conversations by the user U on an online bulletin board site.

[0053] The information processing apparatus 1 inputs, for example, information including instruction information for instructing to select, from among a plurality of impressions IM1 to IMm indicated by impression information, an impression IM held by the user U from the action content, impression information indicating the plurality of impressions IM1 to IMm included in the selected impression group, and the action content, as input information to the generation AI, and causes the generation AI to output information indicating the impression IM presumed to be held by the user U with respect to the target content.

[0054] For example, when the selected impression group is an impression group based on the values of the user U, assume that the action content is indicated by the character string "XXX···XXX". In this case, the information processing apparatus 1 generates instruction information including the information "The character string 'XXX···XXX' is a comment posted by the user regarding the following target content. Please select three or more of the following images presumed to be held by the user U with respect to the following target content from the following values. The output format should be in csv format as follows.\nValue sense, Reason", the impression information, and the target content.

[0055] In addition, the information processing apparatus 1 can also use, for example, the information "Please estimate the score indicating the degree of the impression presumed to be held by the user U with respect to the target content for each of the following values. The score should be in the range of 1 to 10, and the higher the degree of the impression, the larger the value." instead of the information "Please select three or more of the following images presumed to be held by the user U with respect to the target content from the following values.", but is not limited to such an example.

[0056] Also, the output format is not limited to the csv format. Also, the instruction information may include information of a character string indicating a product, a store, etc. specified by the target content instead of the target content. In this case, such information of the character string is used instead of the information of the above-mentioned character string "the following target content". A part of the above-mentioned instruction information can be set as a system message.

[0057] The information processing apparatus 1 can use the content of a plurality of users U as action content including the instruction information regardless of the attributes of the user U, or can use the content of a plurality of users U as action content including the instruction information for each attribute of the user U. When the information processing apparatus 1 includes the content of a plurality of users U in the instruction information for each attribute of the user U, the information processing apparatus 1 can cause the generation AI to output information indicating one or more impressions IM presumed to be held by the user U for each attribute of the user U.

[0058] Also, when the target content is an image, the information processing apparatus 1 inputs, as input information to the multimodal AI, information including instruction information including information such as "the following image is an image posted by the user regarding the following target content" instead of the information of the character string "the comment that 'XXX···XXX' posted by the user regarding the following target content", and can also cause the multimodal AI to output information indicating one or more impressions IM presumed to be held by the user U.

[0059] Also, when the action content includes characters and an image, the information processing apparatus 1 includes the characters and the image in the instruction information as the action content, inputs, as input information to the multimodal AI, information including such instruction information, and can also cause the multimodal AI to output the information.

[0060] Next, a case will be described in which the estimation method and the impression content type received by the information processing apparatus 1 in step S1 are the second estimation method and the first content type. In this case, for each impression IM included in the selected impression group, the information processing apparatus 1 causes the generation AI to generate reference content estimated to be held by the user U for the impression IM, and compares the reference content with the target content to estimate one or more impressions IM held by the user U for the target content.

[0061] The information processing apparatus 1 generates reference content estimated to be held by the user U for each impression IM classified into a plurality in the selected impression group. When the selected impression group is an impression group based on values according to Schwarz's value theory, the plurality of impressions IM classified in the selected impression group are the above-described 10 types of values or 56 types of values.

[0062] For example, the information processing apparatus 1 inputs input information including instruction information, which is information for instructing the generation of reference content estimated to be held by the user U for the impression indicated by the impression IM, to the generation AI, and causes the generation AI to generate reference content. The reference content is, for example, content indicated by at least one of text and an image.

[0063] The information processing apparatus 1 stores fixed instruction information in advance, and inputs information including the fixed instruction information and impression information to the generation AI as instruction information, thereby causing the generation AI to generate reference content for each impression IM. The fixed instruction information is, for example, when the target content is a product introduction text, information of the character string "Post a certain value. Please create 10 product introduction texts that Japanese users are likely to hold for that value. Do not include the posted value in the text too much."

[0064] Also, the impression information is, for example, information of the character string "Value: Social power". As a result, each of the 10 product introduction texts predicted to be held by the user U for the impression IM of social power is generated as reference content.

[0065] When the generative AI is GPT, the stereotyped instruction information is included in the input information (prompt) as an instruction, and the impression information is included in the input information (prompt) as a user message, but is not limited to such an example. For example, the input information may include, as a user message, information including the string "social power as a value" instead of the information of the string "post a certain value. That value" in the stereotyped instruction information.

[0066] After the information processing apparatus 1 generates the reference content for each impression IM, the information processing apparatus 1 compares the target content to be estimated for the impression IM held by the user U with the reference content for each impression IM. The information processing apparatus 1 vectorizes the target content and each reference content for the comparison between the target content and the reference content for each impression IM.

[0067] The vectorization of the content is performed, for example, by embedding (Embedding) using a language model (for example, a model based on a transformer). The vectorized content is represented by, for example, a vector of several hundred dimensions, but is not limited to such an example.

[0068] The embedding by the language model is, for example, an embedding by text-embedding-ada, BERT (Bidirectional Encoder Representations from Transformers) provided by OpenAI (registered trademark), but is not limited to such an example.

[0069] Note that the vectorization of the content is not limited to the embedding by the language model, and for example, the content may be vectorized by Doc2Vec, the average of word embeddings (Word Embedding), etc. For word embeddings, for example, Word2Vec or fastText is used.

[0070] The information processing apparatus 1 classifies the target content into an impression IM corresponding to the reference content whose vector similarity to the target content is equal to or higher than a threshold value. For example, the information processing apparatus 1 can classify the target content into only one impression IM, or can also classify the target content into two or more impression IMs.

[0071] For example, the information processing apparatus 1 can classify the target content into only one impression IM by classifying the target content into the impression IM corresponding to the reference content whose vector similarity to the target content is equal to or higher than the threshold value and whose similarity is the highest.

[0072] Further, when there are two or more reference contents whose vector similarity to the target content is equal to or higher than the threshold value, the information processing apparatus 1 can classify the target content into two or more impression IMs respectively corresponding to these two or more reference contents.

[0073] The vector similarity is the cosine similarity, but it may also be the Jaccard similarity or the like, or may be the reciprocal of the Euclidean distance or the reciprocal of the Manhattan distance or the like. Further, when using the Euclidean distance or the Manhattan distance, the information processing apparatus 1 classifies, for example, the target content whose Euclidean distance or Manhattan distance from the reference content is less than the threshold value into the impression IM corresponding to the reference content.

[0074] Further, the information processing apparatus 1 can also identify the similarity between vectors or classify vectors, and thus the similarity between contents or classify contents, by using an arbitrary method such as learning a classification model using vectors and handler-labeled data corresponding thereto and using the same for classification.

[0075] Further, when the reference content and the target content are images, the information processing apparatus 1 inputs, as input information, information including the reference content, the target content, and instruction information indicating an instruction to output the similarity between the reference content and the target content to the multimodal AI, and can also output the similarity from the multimodal AI.

[0076] In this case, the information processing apparatus 1 classifies, for example, into the impression IM corresponding to the reference content with a similarity generated by the multimodal AI being equal to or higher than the threshold value and having the highest similarity, or the impression IM corresponding to the reference content with a similarity generated by the multimodal AI being equal to or higher than the threshold value.

[0077] Further, when the reference content and the target content are images, the information processing apparatus 1 can also cause the multimodal AI to generate an explanatory text that is a text explaining what images are included in each of the reference content and the target content.

[0078] In this case, the information processing apparatus 1 vectorizes each of the explanatory text of the reference content and the explanatory text of the target content, and calculates the similarity between the vectorized explanatory text of the reference content and the vectorized explanatory text of the target content. The method for calculating the similarity and the group of impressions on the impression IM of the target content are the same as those in the case where the reference content and the target content are texts.

[0079] Note that when the reference content and the target content are images, the information processing apparatus 1 can also calculate the similarity by directly vectorizing each of the reference content and the target content.

[0080] Further, when the reference content and the target content each include text and an image, the information processing apparatus 1 classifies into the impression IM corresponding to the reference content with an integrated score, which is a weighted sum of the similarity between texts and the similarity between images, being equal to or higher than the threshold value, or the impression IM corresponding to the reference content with the integrated score being equal to or higher than the threshold value and being the largest.

[0081] Next, a case will be described in which the estimation method and the impression content type received by the information processing apparatus 1 in step S1 are the second estimation method and the second content type. In this case, the information processing apparatus 1 collects action content, which is content indicating the actions of the user U related to the target content, from the information processing apparatus 4 or the like.

[0082] The action content is content generated by an action of the user U related to the target content. Similar to the case where the estimation method selected in step S1 is the first estimation method, for example, it is posted content such as reviews, comments, and messages posted by the user U regarding the target content.

[0083] For each impression IM included in the selected impression group, the information processing apparatus 1 causes the generation AI to generate reference content estimated to be held by the user U for the impression IM, and compares such reference content with the action content to estimate one or more impressions IM held by the user U for the target content.

[0084] For each impression IM classified into a plurality in the selected impression group, the information processing apparatus 1 generates reference content estimated to be held by the user U for the impression. For example, the information processing apparatus 1 inputs input information including instruction information, which is information instructing the generation of reference content corresponding to the action content, which is content estimated to be held by the user U for the impression indicated by the impression IM, into the generation AI, and causes the generation AI to generate the reference content. The reference content is, for example, content indicated by at least one of text and an image.

[0085] The information processing apparatus 1 stores first fixed instruction information in advance, and inputs information including the fixed instruction information and impression information as instruction information into the generation AI, so as to cause the generation AI to generate reference content for each impression IM. The fixed instruction information is, for example, when the target content is a product introduction text, the information of the character string "I will post a certain value. Please create 10 comments by users who are presumed to have the value by Japanese users. Do not include the posted value as it is in the text too much."

[0086] Also, the impression information is, for example, the information of the character string "Value: Social power". Thereby, each of the 10 comments of users U presumed to have the impression IM of social power is generated as reference content.

[0087] After the information processing apparatus 1 generates reference content for each impression IM, when the estimation method selected in step S1 is the first estimation method, the impression IM held by the user U or the action content to be the estimation target of the held impression IM is compared with the reference content for each impression IM by the same comparison method.

[0088] For example, the information processing apparatus 1 compares the action content to be the estimation target of the impression IM held by the user U with the reference content for each impression IM for each action content, and estimates one or more impression IMs held by the user U for the target content based on the comparison result.

[0089] The information processing apparatus 1 compares the action content to be the estimation target of the impression IM held by the user U with the reference content for each impression IM for each action content. The information processing apparatus 1 vectorizes each action content and each reference content for the comparison between the action content and the reference content for each impression IM.

[0090] The information processing device 1 performs processing to vectorize action content and calculate the similarity with the vectors of each reference content for each action content. Then, for example, the information processing device 1 estimates, as one or more impressions IM that the user U has of the target content, the impression IM corresponding to the reference content whose average value of the vector similarity with each action content is equal to or greater than a threshold value.

[0091] Also, for example, the information processing device 1 estimates, as one or more impressions IM that the user U has of the target content, the impression IM corresponding to the reference content whose median value of the vector similarity with each action content is equal to or greater than a threshold value.

[0092] Subsequently, based on the above-described estimation result, the information processing device 1 generates, as impression content, content including information indicating one or more impressions IM into which the target content has been classified (step S3). For example, the information processing device 1 generates, as impression content, content that maps a plurality of impressions IM1 to IMm classified into a specific impression group and includes information indicating an impression IM that the user U has or is estimated to have of the target content.

[0093] Subsequently, the information processing device 1 provides the impression content generated in step S3 to the operator O (step S4). In step S4, the information processing device 1 provides the impression content generated in step S3 to the operator O, for example, by transmitting the impression content generated in step S3 to the terminal device 3 of the operator O.

[0094] The impression content shown in FIG. 1 is content in which, in content that maps a plurality of values based on Schwartz's value theory as a plurality of impressions IM1 to IMm (in FIG. 1, m = 56), the impressions IM classified as impressions IM that the user U has or has had of the values based on Schwartz's value theory are highlighted. In the example shown in FIG. 1, content showing introduction texts or introduction images of four different models of a certain vehicle type is shown as the target content.

[0095] In the impression content shown in FIG. 1, the automobile of model A is classified into the impressions of "freedom", "independence", and "self - respect" which are the subdivisions of the impression of "self - determination", and the impression of "inner harmony" which is the subdivision of the impression of "universalism".

[0096] Also, in the impression content shown in FIG. 1, the automobile of model B is classified into the impressions of "a life with changes" and "a lively life" which are the subdivisions of the impression of "stimulation", the impressions of "enjoyment of life" and "fun" which are the subdivisions of the impression of "hedonism", and the impression of "sense of belonging" which is the subdivision of the impression of "safety".

[0097] Also, in the impression content shown in FIG. 1, the automobile of model C is classified into the impressions of "freedom" and "goal selection" which are the subdivisions of the impression of "self - determination", the impressions of "enjoyment of life" and "fun" which are the subdivisions of the impression of "hedonism", the impression of "competent" which is the subdivision of the impression of "achievement", and the impressions of "health" and "safety of family" which are the subdivisions of the impression of "safety".

[0098] Also, in the impression content shown in FIG. 1, the automobile of model D is classified into the impressions of "creativity", "freedom", and "goal selection" which are the subdivisions of the impression of "self - determination", the impression of "enjoyment of life" which is the subdivision of the impression of "hedonism", the impression of "influence" which is the subdivision of the impression of "achievement", and the impression of "wealth" which is the subdivision of the impression of "power".

[0099] In this way, the information processing apparatus 1 receives the selection of the impression IM held by the user U or the target content to be the estimation target of the held impression IM and the selection of the impression group which is a group of impressions IM, and uses the generation AI to generate impression content indicating the impression IM that the user U is estimated to hold or has held for the target content among the plurality of impressions IM included in the received impression group. Thereby, the information processing apparatus 1 can reduce the workload required to estimate the impression of the user U for the target content.

[0100] Hereinafter, the configuration of an information processing system including an information processing apparatus 1, a plurality of terminal devices 2, a terminal device 3, and an information processing apparatus 4 that perform such processing will be described in detail.

[0101] [2. Configuration of Information Processing System] FIG. 2 is a diagram showing an example of the configuration of an information processing system according to an embodiment. As shown in FIG. 2, the information processing system 100 according to the embodiment includes an information processing apparatus 1, a plurality of terminal devices 2, a terminal device 3, and an information processing apparatus 4.

[0102] The plurality of terminal devices 2 are used by different users U. The terminal device 3 is, for example, a terminal device of an operator O. The terminal devices 2 and 3 are, for example, notebook PCs (Personal Computers), desktop PCs, smartphones, tablet PCs, or wearable devices. The wearable device is, for example, smart glasses or a smartwatch, but is not limited to such examples.

[0103] The information processing apparatus 4 provides various online services to the user U. For example, the information processing apparatus 4 provides content of various sites such as an EC site, a news site, a store introduction site, an image posting site, a video viewing site, and an online bulletin board site to the user U, but is not limited to such examples.

[0104] The information processing apparatus 4, for example, receives product content such as a product introduction text indicating the content of product introduction from an operator O or the like as posting content on an EC site, and provides such posting content to each user U. Note that the product introduction text can also be referred to as a product description text. In addition, the posting content on the EC site includes, in addition to the product content, comments and evaluations posted on such products.

[0105] In addition, the information processing apparatus 4 receives news articles from an operator O or the like as posted content on a news site, and provides such posted content to each user U. On the news site, the posted content on the news site includes, in addition to news articles, comments posted on such news articles and the like.

[0106] In addition, the information processing apparatus 4 receives store content such as a store introduction text indicating the introduction content of a store from the operator O as posted content on a store introduction site, and provides such posted content to each user U. In addition, the posted content on the store introduction site includes, in addition to the store content, reviews and comments posted on such store. The store introduction site is, for example, a site for each store type, such as a restaurant introduction site.

[0107] In addition, the information processing apparatus 4 receives image content (for example, still image content or moving image content) from an operator O or the like as posted content on an image posting site, and provides such posted content to each user U. In addition, the information processing apparatus 4 receives moving image content, titles and caption texts in the moving image content, etc. as posted content on a moving image viewing site, and provides such posted content to each user U.

[0108] Each of the information processing apparatus 1, the terminal device 2, the terminal device 3, and the information processing apparatus 4 is connected to be communicable with each other by wire or wirelessly via the network N. Note that the information processing system 100 shown in FIG. 2 may include a plurality of information processing apparatuses 1 and the like.

[0109] The network N includes, for example, a WAN (Wide Area Network) such as the Internet and a mobile communication network such as LTE (Long Term Evolution), 4G (4th Generation), and 5G (5th Generation: the 5th generation mobile communication system).

[0110] The terminal devices 2 and 3 can be connected to the network N via a mobile communication network, short-range wireless communication such as Bluetooth (registered trademark), or wireless LAN (Local Area Network), and can communicate with the information processing device 1, the information processing device 4, and the like.

[0111] [3. Configuration of Information Processing Device 1] FIG. 3 is a diagram showing an example of the configuration of the information processing device 1 according to the embodiment. As shown in FIG. 3, the information processing device 1 includes a communication unit 10, a storage unit 11, and a processing unit 12.

[0112] [3.1. Communication Unit 10] The communication unit 10 is realized by, for example, a communication module or a NIC (Network Interface Card). The communication unit 10 is connected to the network N by wire or wirelessly, and transmits and receives information to and from various other devices. For example, the communication unit 10 transmits and receives information to and from each of the terminal device 2, the terminal device 3, and the information processing device 4 via the network N.

[0113] [3.2. Storage Unit 11] The storage unit 11 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 11 includes a user information storage unit 20.

[0114] [3.2.1. User Information Storage Unit 20] The user information storage unit 20 stores user information including information about the user U. FIG. 4 is a diagram showing an example of a user information table stored in the user information storage unit 20 of the information processing device 1 according to the embodiment. As shown in FIG. 4, the user information table stored in the user information storage unit 20 includes items such as "user ID", "attribute information", and "behavior history".

[0115] The "user ID" is identification information for identifying user U. The "attribute information" is the attribute information of user U corresponding to the "user ID", and includes, for example, information on psychographic attributes and demographic attributes. Demographic attributes are, for example, gender, age, place of residence, and occupation, etc., and psychographic attributes are interests such as travel, clothing, cars, religion, etc., lifestyle, thoughts and ideological tendencies, etc.

[0116] The "behavior history" is the behavior history of user U in the online service, and includes information such as search history, browsing history, posting history, and purchase history. The search history is information on search queries used in the past by user U and the content browsed by user U from among the search results. The information on search queries is, for example, information such as search keywords and search phrases.

[0117] The browsing history includes, for example, information indicating the content browsed by user U in the online service, and the posting history includes, for example, information indicating the content (such as reviews and comments, etc.) posted by user U in the past in the online service. The purchase history includes information on the transaction targets with which user U has conducted transactions in the past.

[0118] [3.3. Processing Unit 12] The processing unit 12 is a controller, and is realized, for example, by various programs (corresponding to an example of an information processing program) stored in the storage device inside the information processing device 1 being executed with a RAM, etc. as a work area by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit).

[0119] Further, the processing unit 12 is a controller and may be implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a GPGPU (General Purpose Graphic Processing Unit).

[0120] As shown in FIG. 3, the processing unit 12 includes a reception unit 30, an acquisition unit 31, a generation unit 32, and a provision unit 33, and realizes or executes the functions and operations of information processing described below. Note that the internal configuration of the processing unit 12 is not limited to the configuration shown in FIG. 3, and any other configuration may be used as long as it can perform the information processing described later.

[0121] [3.3.1. Reception Unit 30] The reception unit 30 receives various information and requests from the terminal device 2 or the terminal device 3 via the network N and the communication unit 10. For example, the reception unit 30 receives the selection of the impression IM held by the user U or the target content to be estimated for the held impression IM and the selection of the impression group which is a group of impressions IM.

[0122] For example, it receives an impression content generation request transmitted from the terminal device 3 of the operator O who creates or provides the target content. The impression content generation request includes the target content for which the impression IM held by the user U or the held impression IM is to be estimated, information indicating the impression group which is a group of impressions IM, information specifying the estimation method, information indicating the target content type, and information indicating the impression content type.

[0123] The target content, impression group, estimation method, target content type, and impression content type specified in the impression content generation request are, for example, the target content, impression group, estimation method, target content type, and impression content type selected by the operator O.

[0124] The reception unit 30 receives a request for generating impression content, and thereby receives selections in terms of target content, impression groups, estimation methods, target content types, and impression content types. Note that the operator O is an employee of a company that creates or provides the target content, but may also be an employee of a company commissioned by such a company, etc.

[0125] The target content is web content, for example, a web page, a catch copy, a creative, a stamp, but is not limited to such examples. The target content is, for example, content that is a candidate for or a target for provision on an EC site, a news site, a restaurant introduction site, an image posting site, a video viewing site, etc., and includes advertising content, but is not limited to such examples. The EC site is, for example, a shopping site, an auction site, etc., but is not limited to such examples.

[0126] The target content is content that can be provided on an online site provided by the information processing device 4, and the user U can operate the terminal device 2 to use the online site provided by the information processing device 4. The online site provided by the information processing device 4 is an EC site, a news site, a store introduction site, an image posting site, a video viewing site, etc., but is not limited to such examples.

[0127] The impression group specified in the impression content generation request is an impression group designated by the user U from among a plurality of impression groups C1 to Cn (n is an integer of 2 or more). The plurality of impression groups C1 to Cn are, for example, impression groups classified according to the values of the user U, impression groups classified according to the preferences of the user U, impression groups classified according to the lifestyle of the user U, etc., but are not limited to such examples.

[0128] The impression group according to the values of the user U is, for example, an impression group that classifies impressions IM held or had by the user U as impressions based on the value theory of Schwarz, and each of the above-described 10 classification conditions is further subdivided into a plurality of classification conditions and classified into 56 classification conditions, but is not limited to such examples.

[0129] The values of the user U are not limited to the above examples, and may be, for example, values classified as traditionalism, success orientation, self-actualization orientation, symbiosis orientation, etc., or other values. For example, when the object is a car, the values of the user U may be, for example, values classified as safety-oriented, habitability-oriented, design-oriented, driving performance-oriented, etc. When the object is a house, the values of the user U may be, for example, values classified as location-oriented, price-oriented, floor plan-oriented, design-oriented, future-oriented, etc.

[0130] The impression groups based on the lifestyle of the user U are, for example, impression groups by the AIO approach or impression groups by VALS. The impression groups by the AIO approach are impression groups classified by one or more of impressions such as activities, interests, and opinions. The impression groups by VALS are, for example, impression groups classified as self-actualizers, achievers, belongers, sustainers, strugglers, aspirers, believers, thinkers, and young thinkers.

[0131] Also, the impression groups may be, for example, impression groups shown by the innovator theory that classifies the process of the spread of new products. This impression group is an impression group classified as innovators, early adopters, early majority, late majority, and laggards.

[0132] There are a plurality of estimation methods including a first estimation method and a second estimation method for the method of estimating the impression IM held by the user U for the target content or the estimated impression IM. The first estimation method is a method of directly estimating by the generation AI the impression IM held by the user U for the target content or the impression IM already held. The second estimation method is a method of generating, for each impression IM included in the impression IM group, by the generation AI the reference content estimated to be held by the user U for the impression IM, and comparing such reference content with the target content to estimate the impression IM held by the user U for the target content or the impression IM already held.

[0133] The target content type is, for example, the type of target content, and is indicated by, for example, a combination of an online site and a content type. For example, the target content type includes product descriptions on product pages of e-commerce sites, news articles on news sites, comments on news articles on news sites, image content on image posting sites, title and caption text of video content on video viewing sites, etc., but is not limited to such examples.

[0134] The impression content type is the type of impression content, and includes, for example, a first content type and a second content type. The first content type is the type of impression content that includes information indicating an impression IM that is presumed to be held by the user U with respect to the target content, and is the type of impression content that supports the creation and consideration of the target content. The second content type is the type of impression content that includes information indicating an impression IM that is presumed to have been held by the user U with respect to the target content, and is the type of impression content that supports the grasping of the impression IM that is presumed to have been held by the user U with respect to the target content.

[0135] [[3.3.2. Acquisition Unit 31]] The acquisition unit 31 acquires various information from the information processing device 4 and the storage unit 11. For example, the acquisition unit 31 acquires various contents and information of each user U from the information processing device 4 via the network N and the communication unit 10. Also, the acquisition unit 31 acquires information of each user U from the user information storage unit 20 of the storage unit 11.

[0136] For example, when an impression content generation request is received by the reception unit 30, the acquisition unit 31 acquires, from the impression content generation request, the target content that is the object of estimation of the impression IM held or had by the user U, information indicating an impression group that is a group of impression IMs, information specifying the estimation method, information indicating the target content type, and information indicating the impression content type.

[0137] In addition, when the target content type indicated by the impression content generation request received by the reception unit 30 is the second content type, the acquisition unit 31 collects behavior content, which is content indicating the behavior of the user U regarding the target content, from the information processing device 4 or the like.

[0138] The behavior content is content generated by the behavior of the user U related to the target content. For example, it is posted content such as reviews, comments, and messages posted by the user U regarding the target content. For example, when the target content is a product description or an image of a product, the posted content as behavior content is a review or comment posted on the product introduced by the target content.

[0139] Also, when the target content is a store description or an image inside the store, the posted content as behavior content is a review or comment posted on the store introduced by the target content. Also, when the target content is video content, the posted content as behavior content is a review or comment posted regarding the video content.

[0140] Also, the posted content as behavior content may be content such as conversations and posts by the user U on the official accounts of products and stores indicated by the target content in the communication application. Also, the posted content as behavior content may be content such as posts and conversations by the user U on an online bulletin board site, for example.

[0141] 〔3.3.3. Generation Unit 32〕 The generation unit 32 generates impression content indicating the impression IM that the user U has or is estimated to have had regarding the target content among the plurality of impressions IM included in the impression group received by the reception unit 30, using the generation AI.

[0142] The generative AI used in step S2 is, for example, text generation AI, image generation AI, multimodal AI, etc. Such generative AI is, for example, arranged in an external information processing device, and the generation unit 32 causes the generative AI to generate reference content for each impression via an API provided by the external information processing device and acquires it, but is not limited to such an example. For example, the generative AI may be arranged within the generation unit 32.

[0143] The text generation AI is, for example, a language model trained to estimate and output the next token from the input token sequence, and is, for example, a transformer-based model or an RNN-based model.

[0144] The transformer-based models are, for example, GPT, BARD, etc., but are not limited to such examples. The RNN-based models are, for example, RWKV, etc., but are not limited to such examples. Note that it is desirable to perform learning so that the input information is not used as a new answer to conceal information such as the input personal information.

[0145] The image generation AI is, for example, StackGAN, AttnGAN, T2I (Text-to-Image) with Transformers, DALL-E, etc., but is not limited to such examples. The multimodal AI is, for example, a model that generates an image from text or generates text from an image, and is, for example, GPT-4V, CM3Leon, etc., but is not limited to such examples.

[0146] The generation unit 32 can use a plurality of estimation methods including a second estimation method by a first estimation method as an estimation method for estimating the impression IM held by the user U or the impression IM held. The first estimation method is a method of directly causing the generative AI to estimate the impression IM held by the user U or the impression IM held with respect to the target content.

[0147] The second estimation method is a method of causing the generation AI to generate reference content that is estimated to be held or has been held by the user U for each impression included in the impression group, and comparing such reference content with the target content to estimate the impression IM or the impression IM held by the user U with respect to the target content.

[0148] Here, assume that the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the first estimation method and the first content type. In this case, for example, the generation unit 32 inputs information including instruction information for instructing to select, from among a plurality of impressions IM1 to IMm indicated by impression information, the impression IM held by the user U with respect to the target content, impression information indicating the plurality of impressions IM1 to IMm included in the selected impression group, and the target content, as input information to the generation AI, and causes the generation AI to output information indicating one or more impressions IM estimated to be held by the user U with respect to the target content.

[0149] Also, assume that the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the first estimation method and the second content type. In this case, for example, the generation unit 32 inputs information including instruction information for instructing to select, from among a plurality of impressions IM1 to IMm indicated by impression information, the impression IM held by the user U from the action content, impression information indicating the plurality of impressions IM1 to IMm included in the selected impression group, and the action content, as input information to the generation AI, and causes the generation AI to output information indicating the impression IM estimated to be held by the user U with respect to the target content.

[0150] Also, assume that the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the second estimation method and the first content type. In this case, for each impression IM included in the selected impression group, the generation unit 32 causes the generation AI to generate reference content that is estimated to be held by the user U for the impression IM, and compares such reference content with the target content to estimate one or more impressions IM held by the user U with respect to the target content.

[0151] Also, assume that the estimation method and impression content type indicated by the impression content request received by the reception unit 30 are the second estimation method and the second content type. In this case, for each impression IM included in the selected impression group, the generation unit 32 causes the generation AI to generate reference content estimated to be held by the user U for the impression IM, and compares the reference content with the action content to estimate one or more impressions IM held by the user U for the target content.

[0152] The generation unit 32 includes a first generation processing unit 40, an estimation processing unit 41, and a second generation processing unit 42. The first generation processing unit 40 executes processing when the estimation method indicated by the impression content request received by the reception unit 30 is the second estimation method.

[0153] 〔3.3.3.1. First generation processing unit 40〕 When the estimation method indicated by the impression content request received by the reception unit 30 is the second estimation method, the first generation processing unit 40 generates reference content estimated to be held or held by the user U for each impression IM classified into a plurality in the selected impression group.

[0154] 〔3.3.3.2. Processing in the case of the first content type〕 First, the case where the impression content type indicated by the impression content request received by the reception unit 30 is the first content type will be described. In this case, the first generation processing unit 40 generates reference content estimated to be held by the user U for each impression IM classified into a plurality in the selected impression group.

[0155] For example, the first generation processing unit 40 causes the generation AI to generate reference content estimated to be held by the user U for each impression IM included in the selected impression group. When the selected impression group is an impression group based on values according to Schwarz's value theory, the plurality of impressions IM classified in the selected impression group are the above-described 10 types of values or 56 types of values.

[0156] For example, the first generation processing unit 40 inputs input information including instruction information, which is information instructing the generation of reference content that is presumed to be held by the user U as the impression indicated by the impression IM, into the generation AI, and causes the generation AI to generate the reference content. The reference content is, for example, content indicated by at least one of text and images.

[0157] The first generation processing unit 40 stores fixed instruction information in advance, and by inputting information including the fixed instruction information and the impression information as the instruction information into the generation AI, causes the generation AI to generate reference content for each impression IM. The fixed instruction information is, for example, when the target content is a product introduction text, information in the character string "I will post a certain value. Please create 10 product introduction texts that Japanese users are likely to hold such a value. Do not include the posted value in the text as it is."

[0158] Also, the impression information is, for example, information in the character string "Value: Social power". As a result, each of the 10 product introduction texts predicted to be held by the user U as the impression IM of social power is generated as the reference content.

[0159] When the generation AI is GPT, the fixed instruction information is included in the input information (prompt) as an instruction, and the impression information is included in the input information (prompt) as a user message, but it is not limited to such an example. For example, the input information may include information in the character string "Social power as a value" as the user message instead of the information in the character string "I will post a certain value. That value" in the fixed instruction information.

[0160] 〔3.3.3.3. Processing in the case of the second content type〕 Next, the case where the impression content type indicated by the impression content request received by the reception unit 30 is the second content type will be described. In this case, the first generation processing unit 40 generates reference content that is presumed to be held by the user U for each impression IM classified into a plurality in the selected impression group.

[0161] The first generation processing unit 40 generates reference content that is estimated to be held by the user U for each impression IM classified into a plurality of groups in the selected impression group. For example, the first generation processing unit 40 inputs input information including instruction information, which is information indicating the generation of reference content corresponding to action content and is content estimated to be held by the user U with the impression indicated by the impression IM, into the generation AI, and causes the generation AI to generate the reference content. The reference content is, for example, content indicated by at least one of text and an image.

[0162] The first generation processing unit 40 stores the first fixed instruction information in advance, and inputs information including the fixed instruction information and the impression information as instruction information into the generation AI, thereby causing the generation AI to generate reference content for each impression IM. The fixed instruction information is, for example, when the target content is a product introduction text, information of the character string "Post a certain value. Please create 10 comments by users who are estimated to have the value held by Japanese users. Do not include the posted value as it is too much in the text."

[0163] Also, the impression information is, for example, information of the character string "Value: Social power". As a result, each of the 10 comments of the user U estimated to have the impression IM of social power is generated as reference content.

[0164] 〔3.3.3.4. Estimation processing unit 41〕 The estimation processing unit 41 estimates the impression IM held by the user U for the target content or the impression IM that has been held. The estimation processing unit 41 estimates the impression IM held by the user U for the target content or the impression IM that has been held based on the estimation method and the impression content type indicated in the impression content request received by the reception unit 30.

[0165] 〔3.3.3.4.1. Processing in the case of the first estimation method and the first content type〕 The estimation method indicated by the impression content request received by the reception unit 30 and the impression content type are the first estimation method and the first content type will be described. In this case, the estimation processing unit 41 causes the generation AI to estimate from among a plurality of impressions IM1 to IMm included in the selected impression group the impression IM that the user U has with respect to the target content.

[0166] For example, the estimation processing unit 41 inputs, as input information to the generation AI, information including instruction information instructing to select from among a plurality of impressions IM1 to IMm indicated by impression information the impression IM that the user U has with respect to the target content, impression information indicating a plurality of impressions IM1 to IMm included in the selected impression group, and the target content, and causes the generation AI to output information indicating one or more impressions IM estimated to be had by the user U with respect to the target content.

[0167] FIG. 5 is a diagram showing an example of the input information input to the generation AI by the estimation processing unit 41 in the processing unit 12 of the information processing apparatus 1 according to the embodiment, where the selected impression group is an impression group based on the values of the user U, and the target content is an example in the case of a catchphrase "Because it is used every day, the difference can be noticed."

[0168] In the input information 50 shown in FIG. 5, the estimation processing unit 41 includes, as instruction information in the input information, information of the character string "Please select three or more images that fit the catchphrase 'Because it is used every day, the difference can be noticed.' from the following values. The output format should be in csv format as follows.\nValue sense, Reason".

[0169] Also, in the input information 50, the estimation processing unit 41 includes, as impression information in the input information, information including the character string "Social power, Authority, Wealth, ···, Family security, Reciprocation of favors, Health, Sense of belonging".

[0170] For example, instead of the information in the character string "Please select three or more images that fit the following catchphrase from the following values.", the estimation processing unit 41 can use the information in the character string "Please select three or more impressions that are presumed to be held by the user for the following catchphrase from the following values." or the information in the character string "Please estimate the score indicating the degree of impression that the user is presumed to hold for the following catchphrase for each of the following values. The score should be in the range of 1 to 10, and the higher the degree of impression, the larger the value.", but is not limited to such examples. Also, the output format is not limited to the csv format.

[0171] In addition, the instruction information includes character information corresponding to the target content type. In the above example, since the target content type is a catchphrase, the information in the character string "catchphrase" is included in the instruction information. However, when the target content type is a product description on an EC site product page, the instruction information includes the information in the character string "product description".

[0172] In this way, the estimation processing unit 41 generates instruction information corresponding to the target content type by including character information corresponding to the target content type in the instruction information, but it is also possible to input instruction information that does not depend on the target content type to the generation AI.

[0173] For example, the estimation processing unit 41 can set instruction information including the information in the character string "Please select three or more images that fit the input information from the following values." as a system message instead of the character string "Since it is something used every day, the difference can be noticed. Please select three or more images that fit the following catchphrase from the following values.". The system message is a message that instructs the behavior of the generation AI. For example, in the GPT model of OpenAI, it is a message that is set as the content of "role": "system" in the input information.

[0174] In addition, the estimation processing unit 41 can also estimate the impression IM that the user U has for the target content by limiting the attributes of the user U, etc., using a generation AI selected from among a plurality of impressions IM1 to IMm included in the selected impression group. In this case, for example, instead of the information of the character string "Please select three or more images that fit the following catchphrase from the following values.", the estimation processing unit 41 can use information such as "Please select three or more impressions that are estimated to be held by a 20-year-old male for the following catchphrase from the following values." or the information of the character string "Please estimate the score indicating the degree of the impression that a 20-year-old male has for the following catchphrase for each of the following values. The score should be in the range of 1 to 10, and the higher the degree of the impression, the larger the value." When causing the generation AI to estimate one or more impressions IM held by a user U with an attribute other than a 20-year-old male, it is possible to replace the information of the character string "20-year-old male" with information indicating another attribute.

[0175] In addition, when the target content is image content, the estimation processing unit 41 can input information including the target content and instruction information including information such as the character string "Please select three or more images that fit the input image from the following values." as input information to the multimodal AI, and cause the multimodal AI to output a similarity. Note that the estimation processing unit 41 can limit the attributes of the user U, etc., and output a score indicating the degree of the impression IM, in the same manner as when the target content is other than image content.

[0176] [3.3.3.4.2. Processing in the case of the first estimation method and the second content type] The case where the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the first estimation method and the second content type will be described.

[0177] In this case, for example, the estimation processing unit 41 inputs information including instruction information for instructing to select, from among a plurality of impressions IM1 to IMm indicated by impression information, an impression IM held by the user U from the action content, impression information indicating the plurality of impressions IM1 to IMm included in the selected impression group, and the action content, as input information to the generation AI, and causes the generation AI to output information indicating the impression IM estimated to be held by the user U with respect to the target content.

[0178] For example, when the selected impression group is an impression group based on the values of the user U, assume that the action content is indicated by the character string "XXX···XXX". In this case, the estimation processing unit 41 generates instruction information including information such as "The character string 'XXX···XXX' is a comment posted by the user regarding the following target content. Please select three or more images estimated to be held by the user U regarding the following target content from the following values. The output format should be in csv format as follows.\nValue sense, Reason", impression information, and the target content.

[0179] Further, for example, instead of the information such as "Please select three or more images held by the user U regarding the target content from the following values.", the estimation processing unit 41 can use information such as "Please estimate the score indicating the degree of the impression estimated to be held by the user U regarding the target content for each of the following values. The score should be in the range of 1 to 10, and the higher the degree of the impression, the larger the value.", but is not limited to such an example.

[0180] Also, the output format is not limited to the csv format. Further, the instruction information may include information of a character string indicating a product or store specified by the target content instead of the target content. In this case, such character string information is used instead of the information of the above-mentioned character string "the following target content". A part of the above-mentioned instruction information can be set as a system message.

[0181] The estimation processing unit 41 can use the content of a plurality of users U as action content included in the instruction information regardless of the attributes of the user U, or can use the content of a plurality of users U as action content included in the instruction information for each attribute of the user U. When the estimation processing unit 41 includes the content of a plurality of users U in the instruction information for each attribute of the user U, it can cause the generation AI to output information indicating one or more impressions IM estimated to be held by the user U for each attribute of the user U.

[0182] In addition, when the target content is an image, the estimation processing unit 41 inputs, as input information to the multimodal AI, information including information such as "The following image is an image posted by the user regarding the following target content." instead of the information of the character string "『XXX···XXX』 is a comment posted by the user regarding the following target content.", and can also cause the multimodal AI to output information indicating one or more impressions IM estimated to be held by the user U.

[0183] In addition, when the action content includes characters and an image, the estimation processing unit 41 includes the characters and the image as action content in the instruction information, inputs the information including such instruction information as input information to the multimodal AI, and can also cause the multimodal AI to output.

[0184] 〔3.3.3.4.3. Processing in the case of the second estimation method and the first content type〕 The case where the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the second estimation method and the first content type will be described.

[0185] In this case, the estimation processing unit 41 compares the target content that is the estimation target of the impression IM held by the user U with the reference content for each impression IM, and estimates one or more impressions IM held by the user U for the target content based on the comparison result.

[0186] The estimation processing unit 41 compares the target content that is the object of estimation of the impression IM held by the user U with the reference content for each impression IM. The estimation processing unit 41 vectorizes the target content and each reference content for the comparison between the target content and the reference content for each impression IM.

[0187] The vectorization of the content is performed, for example, by embedding (Embedding) using a language model (for example, a transformer-based model). The vectorized content is represented by a vector of, for example, several hundred dimensions, but is not limited to such an example. The embedding by the language model is, for example, the embedding by text-embedding-ada, BERT, etc. provided by OpenAI (registered trademark), but is not limited to such an example.

[0188] Note that the vectorization of the content is not limited to the embedding by the language model, and the content may be vectorized, for example, by Doc2Vec, the average of word embeddings (Word Embedding), etc. For word embeddings, for example, Word2Vec, fastText, etc. are used.

[0189] The estimation processing unit 41 classifies the target content into the impression IM corresponding to the reference content whose vector similarity with the target content is equal to or higher than the threshold. The estimation processing unit 41 can, for example, classify the target content into only one impression IM, or can also classify the target content into two or more impression IMs.

[0190] For example, the estimation processing unit 41 can classify the target content into only one impression IM by classifying the target content into the impression IM corresponding to the reference content whose vector similarity is equal to or higher than the threshold and whose similarity is the highest.

[0191] In addition, when there are two or more reference contents whose vector similarity to the target content is equal to or greater than a threshold value, the estimation processing unit 41 can classify the target content into two or more impressions IM corresponding to these two or more reference contents respectively.

[0192] The vector similarity is cosine similarity, but it may also be Jaccard similarity or the like, or may be the reciprocal of Euclidean distance or the reciprocal of Manhattan distance. Further, when using Euclidean distance or Manhattan distance, the estimation processing unit 41 classifies, for example, a target content whose Euclidean distance or Manhattan distance from a reference content is less than a threshold value into the impression IM corresponding to the reference content.

[0193] In addition, the estimation processing unit 41 can also use an arbitrary method such as learning a classification model using vectors and handler-labeled data corresponding thereto and classifying using the same, to identify the similarity between vectors, classify vectors, and thus identify the similarity between contents or classify contents.

[0194] In addition, when the reference content and the target content are images, the estimation processing unit 41 inputs, as input information, information including the reference content, the target content, and instruction information indicating an instruction to output the similarity between the reference content and the target content to the multimodal AI, and can also cause the multimodal AI to output the similarity.

[0195] In this case, the estimation processing unit 41 classifies, for example, into the impression IM corresponding to the reference content with the highest similarity whose similarity generated by the multimodal AI is equal to or greater than the threshold value or the impression IM corresponding to the reference content whose similarity generated by the multimodal AI is equal to or greater than the threshold value.

[0196] In addition, when the reference content and the target content are images, the estimation processing unit 41 can also cause the multimodal AI to generate an explanatory text that is a text explaining what images are included in each of the reference content and the target content.

[0197] In this case, the estimation processing unit 41 vectorizes each of the explanatory text of the reference content and the explanatory text of the target content, and calculates the similarity between the vectorized explanatory text of the reference content and the vectorized explanatory text of the target content. The method for calculating the similarity and the group of impressions on the impression IM of the target content are the same as those in the case where the reference content and the target content are texts.

[0198] Note that when the reference content and the target content are images, the estimation processing unit 41 can also calculate the similarity by directly vectorizing each of the reference content and the target content.

[0199] In addition, when the reference content and the target content each include text and an image, the estimation processing unit 41 classifies the impression IM corresponding to the reference content whose integrated score, which is the weighted sum of the similarity between texts and the similarity between images, is equal to or higher than the threshold value, or the impression IM corresponding to the reference content whose integrated score is equal to or higher than the threshold value and is the largest.

[0200] 〔3.3.3.4.4. Processing in the case of the second estimation method and the second content type〕 Next, the case where the estimation method and the impression content type indicated by the impression content request received by the reception unit 30 are the second estimation method and the second content type will be described.

[0201] In this case, the estimation processing unit 41 compares the action content that is the estimation target of the impression IM held by the user U with the reference content for each impression IM for each motion content, and based on the comparison result, estimates one or more impressions IM held by the user U for the target content.

[0202] The estimation processing unit 41 compares, for each behavior content, the behavior content that is the estimation target of the impression IM held by the user U with the reference content for each impression IM. For the comparison between the behavior content and the reference content for each impression IM, the estimation processing unit 41 vectorizes each behavior content and each reference content. The method of vectorizing the content is as described above.

[0203] The estimation processing unit 41 performs, for each behavior content, a process of calculating the similarity between the vectorized behavior content and the vector of each reference content. Then, for example, the estimation processing unit 41 estimates, as one or more impressions IM held by the user U with respect to the target content, the impression IM corresponding to the reference content whose average value of the vector similarity with each behavior content is equal to or greater than the threshold value.

[0204] Also, for example, the estimation processing unit 41 estimates, as one or more impressions IM held by the user U with respect to the target content, the impression IM corresponding to the reference content whose median value of the vector similarity with each behavior content is equal to or greater than the threshold value.

[0205] 〔3.3.3.5. Second generation processing unit 42〕 Based on the estimation result by the estimation processing unit 41, the second generation processing unit 42 generates, as impression content, content including information indicating one or more impressions IM into which the target content is classified.

[0206] For example, the second generation processing unit 42 generates, as impression content, content that maps a plurality of impressions IM1 to IMm classified into a specific impression group and includes information indicating the impression IM that the user U is supposed to hold or has held with respect to the target content.

[0207] 〔3.3.4. Provision unit 33〕 The providing unit 33 provides the impression content generated by the generating unit 32. For example, the providing unit 33 provides the impression content generated by the generating unit 32 to the operator O by transmitting, for example, the impression content generated by the generating unit 32 to the terminal device 3 of the operator O via the communication unit 10 and the network N.

[0208] The terminal device 3 receives the impression content transmitted from the providing unit 33 of the information processing device 1 via the communication unit 10 and the network N, and displays the received impression content. FIG. 6 is a diagram showing an example of the impression content generated by the second generation processing unit 42 in the processing unit 12 of the information processing device 1 according to the embodiment.

[0209] The impression content 60 shown in FIG. 6 is content in which a plurality of values based on Schwartz's value theory are mapped as a plurality of impressions IM1 to IMm (in FIG. 1, m = 56), and the impression IM classified as the impression held by the user U based on the values based on Schwartz's value theory is highlighted. In the example shown in FIG. 1, content showing introduction texts or introduction images of four different models of a certain vehicle type is shown as the target content.

[0210] In the impression content 60, the vehicle of model A is classified into the impressions "freedom", "independence", and "self - respect" which are the subdivisions of the impression "self - determination", and the impression "inner harmony" which is the subdivision of the impression "universalism".

[0211] Also, in the impression content 60, the vehicle of model B is classified into the impressions "a life with changes" and "a lively life" which are the subdivisions of the impression "stimulation", the impressions "enjoyment of life" and "fun" which are the subdivisions of the impression "hedonism", and the impression "sense of belonging" which is the subdivision of the impression "security".

[0212] In addition, in the impression content 60, automobiles of type C are classified into the impression "freedom" and the impression "goal selection" which are subdivisions of the impression "self-determination", the impression "enjoyment of life" and the impression "fun" which are subdivisions of the impression "hedonism", the impression "competent" which is a subdivision of the impression "achievement", and the impression "health" and the classification "safety of family" which are subdivisions of the impression "safety".

[0213] In addition, in the impression content 60, automobiles of type D are classified into the impression "creativity", the impression "freedom", and the impression "goal selection" which are subdivisions of the impression "self-determination", the impression "enjoyment of life" which is a subdivision of the impression "hedonism", the impression "influence" which is a subdivision of the impression "achievement", and the impression "wealth" which is a subdivision of the impression "power".

[0214] [4. Processing Procedure] Next, the information processing procedure by the processing unit 12 of the information processing apparatus 1 according to the embodiment will be described. FIG. 7 is a flowchart showing an example of information processing by the processing unit 12 of the information processing apparatus 1 according to the embodiment.

[0215] As shown in FIG. 7, the processing unit 12 of the information processing apparatus 1 determines whether an impression content request has been received (step S10). When the processing unit 12 determines that an impression content request has been received (step S10: Yes), it estimates, using the generation AI, the impression that the user U is considered to have or has had based on the information included in the impression content request (step S11).

[0216] Subsequently, the processing unit 12 generates impression content based on the impression estimated in step S11 (step S12). Then, the processing unit 12 provides the impression content generated in step S12 to the operator O (step S13).

[0217] When the process in step S13 ends, or when it is determined that no impression content request has been received (step S10: No), the processing unit 12 determines whether the operation end timing has arrived (step S14). For example, the processing unit 12 determines that the operation end timing has arrived when the power of the information processing apparatus 1 is turned off.

[0218] When the processing unit 12 determines that the operation end timing has not arrived (step S14: No), the process proceeds to step S10. When the processing unit 12 determines that the operation end timing has arrived (step S14: Yes), the process shown in FIG. 7 ends.

[0219] 5. Modification Example In the above-described example, the generation unit 32 estimates the impression IM held by the user U based on the action content of the user U. However, based on the attribute information (for example, psychographic attributes, etc.) of the user U who has performed an action related to the target content and other action information, it is also possible to estimate the impression IM held by the user U with respect to the target content.

[0220] Further, the generation unit 32 can also estimate the impression IM held or had by the user U with respect to the target content by weighted addition of the impression IM estimated by the first estimation method and the impression IM estimated by the second estimation method.

[0221] Further, the generation unit 32 can also estimate, as the impression IM held or had by the user U with respect to the target content, the impression IM whose result of weighted addition of the score of each impression IM estimated by the first estimation method and the score of each impression IM estimated by the second estimation method is equal to or greater than a threshold value.

[0222] Further, the generation unit 32 can also estimate, as the impression IM held or had by the user U with respect to the target content, the impression IM whose estimation results by the first estimation method and the second estimation method match.

[0223] Further, the generation unit 32 can also estimate a plurality of impressions IM including the impression IM estimated by the first estimation method and the impression IM estimated by the second estimation method as the impression IM that the user U has or has had with respect to the target content.

[0224] [6. Hardware Configuration] The information processing apparatus 1 according to the above-described embodiment is realized by a computer 80 having a configuration as shown in FIG. 8, for example. FIG. 8 is a hardware configuration diagram showing an example of a computer 80 that realizes the functions of the information processing apparatus 1 according to the embodiment. The computer 80 includes a CPU 81, a RAM 82, a ROM (Read Only Memory) 83, an HDD (Hard Disk Drive) 84, a communication interface (I / F) 85, an input / output interface (I / F) 86, and a media interface (I / F) 87.

[0225] The CPU 81 operates based on a program stored in the ROM 83 or the HDD 84 and controls each part. The ROM 83 stores a boot program executed by the CPU 81 when the computer 80 is started up, and programs dependent on the hardware of the computer 80.

[0226] The HDD 84 stores programs executed by the CPU 81 and data used by such programs. The communication interface 85 receives data from other devices via the network N (see FIG. 2) and sends it to the CPU 81, and sends data generated by the CPU 81 to other devices via the network N.

[0227] The CPU 81 controls output devices such as a display and a printer, and input devices such as a keyboard or a mouse via the input / output interface 86. The CPU 81 acquires data from the input device via the input / output interface 86. Further, the CPU 81 outputs data generated via the input / output interface 86 to the output device.

[0228] The media interface 87 reads a program or data stored in the recording medium 88 and provides it to the CPU 81 via the RAM 82. The CPU 81 loads such a program from the recording medium 88 onto the RAM 82 via the media interface 87 and executes the loaded program. The recording medium 88 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0229] For example, when the computer 80 functions as the information processing apparatus 1 according to the embodiment, the CPU 81 of the computer 80 realizes the functions of the processing unit 12 by executing the program loaded onto the RAM 82. Further, the HDD 84 stores the data in the storage unit 11. The CPU 81 of the computer 80 reads and executes these programs from the recording medium 88, but as another example, these programs may be acquired from another device via the network N.

[0230] 〔7. Others〕 Also, among the respective processes described in the above embodiment, all or part of the processes described as being automatically performed can be manually performed, or all or part of the processes described as being manually performed can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

[0231] Also, each component of each device shown in the drawings is a functional concept, and it is not necessarily physically configured as shown in the drawings. That is, the specific form of the dispersion and integration of each device is not limited to that shown in the drawings, and all or part of it can be functionally or physically dispersed and integrated in any unit according to various loads and usage situations.

[0232] For example, the above-described information processing apparatus 1 may be implemented by a terminal device and a server computer, or may be implemented by a plurality of server computers. Also, depending on the function, the configuration can be flexibly changed, such as by calling an external platform or the like using an API or network computing.

[0233] Also, the above-described embodiments and modified examples can be appropriately combined as long as the processing contents do not conflict.

[0234] 〔8. Effect〕 As described above, the information processing apparatus 1 according to the embodiment generates a reception unit 30, a generation unit 32, and a provision unit 33. The reception unit 30 receives a selection of the impression IM held by the user U or the target content to be estimated for the held impression IM and a selection of an impression group which is a group of impressions IM. The generation unit 32 generates impression content indicating the impression IM estimated to be held or held by the user U for the target content among the plurality of impressions IM included in the impression group received by the reception unit 30, using a generation AI. The provision unit 33 provides the impression content generated by the generation unit 32. Thereby, the information processing apparatus 1 can reduce the workload required to estimate the impression of the user U with respect to the target content.

[0235] Also, the generation unit 32 causes the generation AI to estimate the impression IM estimated to be held or held by the user U for the target content from among the plurality of impressions IM. Thereby, the information processing apparatus 1 can reduce the workload required to estimate the impression of the user U with respect to the target content.

[0236] Further, the generation unit 32 inputs, as input information to the generation AI, information including impression information indicating a plurality of impressions IM included in the impression group and instruction information instructing to select, from the plurality of impressions IM indicated by the impression information, the impression IM that the user U has or had with respect to the target content, and causes the generation AI to output information indicating the impression IM estimated that the user U has or had with respect to the target content. Thereby, the information processing apparatus 1 can accurately estimate the impression of the user U with respect to the target content.

[0237] Further, the generation unit 32 inputs, as input information to the generation AI, information further including the target content, and causes the generation AI to output information indicating the impression IM estimated that the user U has with respect to the target content. Thereby, the information processing apparatus 1 can accurately estimate the impression of the user U with respect to the target content.

[0238] Further, the generation unit 32 inputs, as input information to the generation AI, action content that is content generated by an action of the user U related to the target content or information further including the target content, and causes the generation AI to output information indicating the impression estimated that the user U has had with respect to the target content. Thereby, the information processing apparatus 1 can accurately estimate the impression of the user U with respect to the target content.

[0239] Further, the generation unit 32 includes a first generation processing unit 40, an estimation processing unit 41, and a second generation processing unit 42. The first generation processing unit 40 generates, for each impression IM included in the impression group, reference content estimated that the user U has or had the impression IM. The estimation processing unit 41 estimates the impression IM estimated that the user U has or had based on the comparison result of comparing the action content that is content generated by an action of the user U related to the target content or the target content with each reference content. The second generation processing unit 42 generates impression content based on the estimation result by the estimation processing unit 41. Thereby, the information processing apparatus 1 can accurately estimate the impression of the user U with respect to the target content.

[0240] In addition, the reception unit 30 receives the selection of one impression group from among a plurality of predetermined impression groups. Thereby, the information processing apparatus 1 can reduce the workload required to estimate the impression of the user U on the target content.

[0241] In addition, the generation unit 32 generates content obtained by mapping a plurality of impressions included in the impression group received by the reception unit 30 as impression content, the content including information indicating the impression IM that the user U has or is estimated to have regarding the target content. Thereby, the information processing apparatus 1 can enable the operator O to easily grasp the impression of the user U on the target content.

[0242] As described above, the embodiments of the present application have been described in detail with reference to the drawings. However, this is an example, and the present invention can be implemented in other forms with various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the column of the disclosure of the invention.

[0243] In addition, the “section (section, module, unit)” described above can be read as “means” or “circuit”. For example, the acquisition unit can be read as an acquisition means or an acquisition circuit.

Explanation of Reference Numerals

[0244] 1,4 Information processing apparatus 2,3 Terminal device 10 Communication unit 11 Storage unit 12 Processing unit 20 User information storage unit 30 Reception unit 31 Acquisition unit 32 Generation unit 33 Provision unit 40 First generation processing unit 41 Estimation processing unit 42 Second generation processing unit 100 Information processing system N Network

Claims

1. A reception unit that receives selection of target content to be estimated for an impression held or had by a user and selection of an impression group that is a group of impressions; A generation unit that generates impression content indicating an impression that the user is estimated to hold or have held for the target content among a plurality of impressions included in the impression group received by the reception unit, using a generation AI; And a provision unit that provides the impression content generated by the generation unit. An information processing apparatus characterized by the above.

2. The generation unit Causes the generation AI to estimate, from among the plurality of impressions, an impression that the user is estimated to hold or have held for the target content. The information processing apparatus according to claim 1, characterized by the above.

3. The generation unit Inputs information including impression information indicating the plurality of impressions included in the impression group and instruction information instructing to select, from among the plurality of impressions indicated by the impression information, an impression that the user is estimated to hold or have held for the target content, as input information to the generation AI, and causes the generation AI to output information indicating an impression that the user is estimated to hold or have held for the target content. The information processing apparatus according to claim 2, characterized by the above.

4. The generation unit Inputs information further including the target content as the input information to the generation AI, and causes the generation AI to output information indicating an impression that the user is estimated to hold for the target content. The information processing apparatus according to claim 3, characterized by the above.

5. The generation unit Inputs, as the input information to the generation AI, action content that is content generated by an action by the user and related to the target content or information further including the target content, and causes the generation AI to output information indicating an impression that the user is estimated to have held for the target content. The information processing apparatus according to claim 3, characterized by the above.

6. The generation unit For each impression included in the impression group, a first generation processing unit that generates reference content for which it is estimated that the user holds or has held the impression. An estimation processing unit that estimates an impression that the user has or is presumed to have had based on behavior content that is content generated by the user's behavior related to the target content or based on a comparison result of comparing the target content with each of the reference contents; A second generation processing unit that generates the impression content based on the estimation result by the estimation processing unit. The information processing apparatus according to any one of claims 1 to 5, characterized in that.

7. The reception unit Receives a selection of one impression group from a plurality of predetermined impression groups. The information processing apparatus according to any one of claims 1 to 5, characterized in that.

8. The generation unit Generates the impression content as content that maps the plurality of impressions included in the impression group received by the reception unit and includes information indicating an impression that the user has or is presumed to have had with respect to the target content. The information processing apparatus according to any one of claims 1 to 5, characterized in that.

9. An information processing method executed by a computer, A reception step of receiving a selection of target content to be an estimation target of an impression that the user has or has had and a selection of an impression group that is a group of impressions; A generation step of generating, using a generation AI, impression content indicating an impression that the user has or is presumed to have had with respect to the target content among the plurality of impressions included in the impression group received in the reception step; An information processing method including a provision step of providing the impression content generated in the generation step. Characterized by that.

10. A reception procedure for receiving a selection of target content to be an estimation target of an impression that the user has or has had and a selection of an impression group that is a group of impressions; A generation procedure for generating, using a generation AI, impression content indicating an impression that the user has or is presumed to have had with respect to the target content among the plurality of impressions included in the impression group received in the reception procedure; Causes a computer to execute a provision procedure for providing the impression content generated in the generation procedure. An information processing program characterized by that.

Images