Information processing apparatus, 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 improving convenience on a providing side of a specific object.SOLUTION: An information processing apparatus includes a target-related information acquisition unit, a user information acquisition unit, and a generation unit. The target-related information acquisition section acquires target information including information on a search target and target impression information indicating an impression that a user is estimated to have about the search target. The user information acquisition unit acquires user impression information indicating an impression that the user is determined to have about the target information. The generation part generates improvement information including information showing improvement contents related to the object information on the basis of the object information and the object impression information acquired by the object related information acquisition part and the user impression information acquired by the user information acquisition part.SELECTED DRAWING: Figure 3

Description

[Technical Field]

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

[0002] Conventionally, attempts have been made to create information on specific objects such as products and services so that the impression a user has of the specific object matches the target impression, but there is a possibility that the impression the user actually has of the specific object may be different.

[0003] Patent document 1 discloses a technology in which an evaluator inputs the evaluation level of the impression they received when viewing content, which is information about a specific target, and the pair of the content and the evaluation level of the impression the evaluator received about this content is used as training data for training a machine learning model. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-133455 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-described conventional techniques are limited to estimating the impression that a user has, and there is room for further improvement in terms of improving the convenience for the provider of the specific target.

[0006] The present application has been made in view of the above, and aims to provide an information processing device, an information processing method, and an information processing program that can improve the convenience of the provider of a specific target. [Means for solving the problem]

[0007] The information processing device according to the present application includes a target-related information acquisition unit, a user information acquisition unit, and a generation unit. The target-related information acquisition unit acquires target information including information about a specific target and target impression information indicating an impression that a user is estimated to have of the specific target. The user information acquisition unit acquires user impression information indicating an impression that the user is determined to have of the target information. The generation unit generates improvement information including information indicating improvement details regarding the target information, based on the target information and target impression information acquired by the target-related information acquisition unit and the user impression information acquired by the user information acquisition unit. [Effects of the Invention]

[0008] According to one aspect of the embodiment, it is possible to achieve an effect of improving convenience for the provider of the identified target. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of an information processing system according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of a user information table stored in the user information storage unit of the information processing device according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of merge information generated by a generating unit in the processing unit of the information processing device according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of template information used by a generating unit in the processing unit of the information processing device according to the embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a prompt generated by a generating unit in the processing unit of the information processing apparatus according to the embodiment using the template information shown in FIG. [Figure 8]FIG. 8 is a diagram showing another example of template information used by the generating unit in the processing unit of the information processing device according to the embodiment. [Figure 9] FIG. 9 is a diagram showing an example of a prompt generated by a generating unit in the processing unit of the information processing apparatus according to the embodiment using the template information shown in FIG. [Figure 10] FIG. 10 is a diagram illustrating an example of before-and-after improvement impression information provided by a providing unit in the processing unit of the information processing device according to the embodiment. [Figure 11] FIG. 11 is a flowchart showing an example of information processing by the processing unit of the information processing device according to the embodiment. [Figure 12] FIG. 12 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, modes for implementing an information processing device, an information processing method, and an 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 device, the information processing method, and the information processing program according to the present application are not limited to these embodiments. Furthermore, the respective embodiments can be appropriately combined within the scope of not causing any contradiction in the processing content. Furthermore, the same components in the following embodiments will be assigned the same reference numerals, and redundant explanations will be omitted.

[0011] [1. An example of information processing] FIG. 1 is a diagram showing an example of information processing according to an embodiment, and in this embodiment, an information processing method is executed by an information processing device 1.

[0012] 1, the information processing device 1 receives an improvement request transmitted from the terminal device 3 of a worker O who creates or provides target information, which is information of a specific target (step S1). The improvement request includes, for example, target information including the information of the specific target, and improvement direction information indicating a direction for improving the target information.

[0013] The specific target may be, for example, a product or a service, but may also be an organization such as a company, a facility such as a school or a hospital, a local public body such as a city, town, or village, or other target. The specific information may be an advertisement for promoting the specific target, such as a catchphrase, package, description, or introduction of the specific target, but is not limited to such examples. For example, if the specific target is video content or music content, the information about the specific target may be the video content or music content itself.

[0014] The direction of improvement of the target information is, for example, a direction of improvement toward the impression that user U is estimated to have, or a direction of improvement toward the impression that user U is determined to have had, but is not limited to such examples.

[0015] For example, the direction of improvement of the target information may be an intermediate direction of improvement between the impression that user U is estimated to have and the impression that user U is determined to have had, or it may be an improvement direction that indicates the degree to which the impression that user U is estimated to have had (e.g., closer to 70%), or it may be an improvement direction that indicates the degree to which the impression that user U is determined to have had (e.g., closer to 80%).

[0016] The improvement request also includes user identification information, which is information for identifying the target user U. The target user U is the user U to whom the target information has been provided, but is not limited to such an example, and may be, for example, the user U to whom the target information has been provided and who has purchased or used a specific target. The target user U may also be the user U to whom the target information has been provided. The user identification information is, for example, information indicating the attributes of the target user U, information indicating the behavior of the target user U, which is the behavior of the user U regarding a specific target, etc., but is not limited to such an example.

[0017] Examples of behaviors related to a specific target include, but are not limited to, posting a review about the specific target and responding to a questionnaire about the specific target by the user U. In addition, behaviors related to a specific target may include evaluations of the specific target posted by the user U (e.g., positive evaluations, negative evaluations, or the degree thereof), browsing behaviors of the web pages of the specific target by the user U, and search behaviors of the specific target by the user U. In addition, behaviors related to a specific target may include comments about the specific target on a social networking service (SNS).

[0018] The improvement request may also include estimation method type information indicating the type of estimation method for the impression that the user U has of the specific object. There are a plurality of estimation methods including a first estimation method and a second estimation method for the type of estimation method for the impression that the user U has of the specific object.

[0019] The first estimation method is a method in which a generation AI (Artificial Intelligence) directly estimates the impression that a user U has of a specific object. The second estimation method is a method in which, for each impression included in an impression group, a generation AI generates reference content that is estimated to have the impression that the user U has, and compares the reference content with the target content to estimate the impression that the user U has of the target content.

[0020] The improvement request may also include determination method type information indicating the type of method for determining the impression the user U has of the specific target. There are multiple types of determination methods for the impression the user U has of the specific target, including a first determination method and a second determination method. The receiving unit 30 receives the determination method type information by receiving the improvement request.

[0021] The first determination method is a method in which the generation AI directly determines the impression that user U had of a specific target. The second estimation method is a method in which, for each impression included in the impression group, the generation AI generates reference content that is determined to be the impression that user U had, and compares the reference content with the target content to determine the impression that user U had of the target content.

[0022] Next, the information processing device 1 acquires user behavior information from the information processing device 4 based on the user identification information included in the improvement request (step S2). The user behavior information is information about the target user U identified by the user identification information, and is information indicating the behavior of the user U regarding a specific target for which content is provided on an online site provided by the information processing device 4.

[0023] The online sites provided by the information processing device 4 include, but are not limited to, EC (Electronic Commerce) sites, news sites, store introduction sites, image posting sites, video viewing sites, etc. The user U can use the online sites provided by the information processing device 4 by operating the terminal device 2.

[0024] The content for a specific target provided on the online site provided by the information processing device 4 includes content for the user U to perform actions related to the specific target described above, such as content including content for posting reviews about the specific target, content including content for answering questionnaires about the specific target, etc., but is not limited to such examples.

[0025] The specific target content includes, but is not limited to, specific information, and may not include specific information. The target information may be, but is not limited to, an advertisement including one or more of a package of the specific target, a catchphrase of the specific target, and a description of the specific target. For example, if the specific target is content, the specific information may be information about the content itself.

[0026] Information about user U identified by the user identification information, which indicates the behavior of user U regarding content provided on the online site provided by the information processing device 4, includes, as described above, for example, reviews posted about specific targets, responses by user U to questionnaires about specific targets, etc., but is not limited to such examples.

[0027] Next, the information processing device 1 estimates the impression held by the user U based on the target information included in the improvement request received in step S1 (step S3). If the improvement request received in step S1 includes estimation method type information, the information processing device 1 estimates the impression held by the user U using the estimation method of the type indicated by the estimation method type information. If the improvement request received in step S1 does not include estimation method type information, the information processing device 1 estimates the impression held by the user U using the first estimation method.

[0028] The generative AI may be, for example, a text generation AI or a multimodal generation AI. The text generation AI is, for example, a large-scale language model trained to estimate and output the next token from an input token sequence, such as a transformer-based model or an RNN (Recurrent Neural Network)-based model, or may be a hybrid model of these. The text generation AI may also be a composite system combined with an identifier to prevent fraudulent use.

[0029] Examples of the transformer-based model include, but are not limited to, GPT (Generative Pre-trained Transformer) (registered trademark), PaLM2 (Pathways Language Model Version 2), and LLaMA (Large Language Model Meta AI). Examples of the RNN-based model include, but are not limited to, RWKV (Receptance Weighted Key Value).

[0030] It is desirable that the generation AI be trained so as not to include personal information in the generated results. The generation AI is placed in an external information processing device, and the information processing device 1 uses the generation AI via an API, but the generation AI may also be placed within the information processing device 1.

[0031] Multimodal generative AI is, for example, generative AI that can generate text and images from text, images, etc. Examples of multimodal generative AI include GPT-4o, gemini, Claude3, and CM3Leon (Chameleon Multimodal Model), but are not limited to these examples.

[0032] First, the first estimation method will be described. The information processing device 1 estimates the impression that the user U has of a specific target from among the multiple impressions included in the impression group using a generation AI, based on the target information included in the improvement request received in step S1. In the following description, it is assumed that the multiple impressions included in the impression group are multiple impressions IM1 to IMm (m is an integer equal to or greater than 2).

[0033] First, the information processing device 1 causes the generation AI to estimate the impression IM that the user U has of a specific target from among the multiple impressions IM1-IMm included in the impression group. For example, the information processing device 1 inputs, as input information to the generation AI, information including instruction information instructing the user to select the impression IM that the user U has of the specific target from the multiple impressions IM1-IMm indicated by impression information, impression information indicating the multiple impressions IM1-IMm included in the impression group, and target information, and causes the generation AI to output information indicating one or more impressions IM that are estimated to be held by the user U of the specific target.

[0034] For example, suppose the impression group is an impression group based on the values ​​of user U, and the target information is a catchphrase, "You'll know the difference because it's something you use every day." In this case, the information processing device 1 includes, for example, a character string, "Please select three or more images from the values ​​below that fit the catchphrase, 'You'll know the difference because it's something you use every day.'\nValues, Reason," as instruction information in the input information. In this way, the instruction information includes the target information included in the improvement request received in step S1.

[0035] Furthermore, the information processing device 1 includes information including the character string "social power, authority, wealth, saving face, social recognition\n success, competence, ambition, influence, intelligence\n fun, enjoying life\n courage, a varied life, a vibrant life\n creativity, curiosity, freedom, choosing goals, self-respect, independence\n environmental protection, the world of beauty, harmony with nature, generosity, social justice, wisdom, equality, world peace, inner harmony\n assistance, honesty, tolerance, loyalty, responsibility, true friendship, the spiritual world, mature love, the meaning of life\n politeness, respect for parents and elders, self-discipline, obedience\n piety, acceptance of fate, humility, moderation, respect for tradition, detachment\n cleanliness, national security, social order, family security, giving back, health, a sense of belonging" in the input information as impression information.

[0036] Furthermore, the information processing device 1 can use, for example, information such as "Please select three or more of the values ​​below that represent the impression that you think the user will have of the catchphrase," instead of the information in the string "Please select three or more of the values ​​below that represent the impression that you think the user will have of the catchphrase," or information such as "Please estimate a score indicating the degree of impression that you think the user will have of the catchphrase for each of the values ​​below. The score should be in the range of 1 to 10, and the higher the value, the higher the degree of impression," but is not limited to such examples.

[0037] Furthermore, the information processing device 1 can also use a generation AI to estimate the impression that the user U has of the target information from multiple impressions IM1 to IMm included in the impression group by limiting the attributes of the user U. In this case, instead of the information string "Please select three or more images from the values ​​below that fit the catchphrase," the information processing device 1 can use information such as "Please select three or more impressions from the values ​​below that are estimated to be formed by a man in his twenties for the catchphrase." or information such as the information string "Please estimate, for each of the values ​​below, a score indicating the degree of impression that a man in his twenties is estimated to have for the catchphrase. The score should range from 1 to 10, with a higher value indicating a higher degree of impression." When the generation AI is to estimate one or more impressions IM formed by a user U with an attribute other than a man in his twenties, the information string "man in his twenties" can be replaced with information indicating another attribute.

[0038] Furthermore, when the target information is image information, the information processing device 1 can input information including the target information, instruction information including information such as a string of characters "Please select three or more images from the values ​​below that fit the input image," and impression information as input information to the multimodal generation AI, and cause the multimodal generation AI to output a similarity. Note that the information processing device 1 can limit the attributes of the user U and output a score indicating the degree of impression IM, just as when the target information is other than image information.

[0039] Next, the second estimation method will be described. The information processing device 1 causes the generation AI to generate reference information that estimates the impression IM that the user U will have for each impression IM included in the impression group, and compares the reference information with the target information to estimate one or more impressions IM that the user U will have for the target information.

[0040] For each impression IM included in the impression group, the information processing device 1 generates reference information that is estimated to indicate the impression that the user U will have. When the impression group is an impression group based on values ​​based on Schwartz's value theory, the multiple impressions IM classified into the impression group are the 10 types of values ​​or 56 types of values ​​described above.

[0041] For example, the information processing device 1 inputs input information including instruction information, which is information instructing the generation AI to generate reference information that is estimated to give the impression indicated by the impression IM to the user U, and causes the generation AI to generate the reference information. The reference information is, for example, information indicated by at least one of text and an image.

[0042] The information processing device 1 stores standard instruction information in advance, and inputs information including the standard instruction information and information about the impression IM to the generation AI as instruction information, causing the generation AI to generate reference information for each impression IM. For example, if the specific object is a "car" and the target information is a catchphrase for the specific object, the standard instruction information is the character string "I will post a certain value. Please create 10 catchphrases for cars that are likely to share the value. Please do not include too many of the posted values ​​directly in the sentences."

[0043] Moreover, the information on the impression IM is, for example, information on the character string “values: social power.” As a result, each of the 10 catchy slogans that are predicted to give the user U the impression IM of social power is generated as reference information.

[0044] After generating the reference information for each impression IM, the information processing device 1 compares the target information that is the target of estimating the impression IM held by the user U with the reference information for each impression IM. In order to compare the target information with the reference information for each impression IM, the information processing device 1 vectorizes the target information and each piece of reference information.

[0045] The vectorization of information is performed, for example, by embedding using a language model (for example, a transformer-based model). The vectorized information is represented, for example, by a vector with several hundred dimensions, but is not limited to such an example.

[0046] Embedding using a language model is, for example, embedding using text-embedding-ada provided by OpenAI (registered trademark) or BERT (Bidirectional Encoder Representations from Transformers), but is not limited to these examples.

[0047] Note that the vectorization of information is not limited to embedding using a language model, and may be performed using, for example, Doc2Vec, the average of word embedding, etc. Word2Vec, fastText, etc. are used for word embedding.

[0048] The information processing device 1 classifies the target information into an impression IM corresponding to reference information whose vector similarity with the target information is equal to or greater than a threshold value. The information processing device 1 can classify the target information into only one impression IM, or into two or more impression IMs.

[0049] For example, the information processing device 1 can classify the target information into only one impression IM by classifying the target information into an impression IM corresponding to the reference information whose vector similarity is equal to or greater than a threshold and has the highest similarity.

[0050] Furthermore, when there are two or more pieces of reference information whose vector similarity with the target information is equal to or greater than a threshold, the information processing device 1 can classify the target information into two or more impressions IM corresponding to the two or more pieces of reference information, respectively.

[0051] The similarity between the vectors is cosine similarity, but may also be Jaccard similarity, or the inverse of the Euclidean distance or the inverse of the Manhattan distance. When the Euclidean distance or the Manhattan distance is used, the information processing device 1 classifies, for example, object information whose Euclidean distance or the Manhattan distance from the reference information is less than a threshold into an impression IM corresponding to the reference information.

[0052] In addition, the information processing device 1 can use any method to learn a classification model using vectors and the corresponding hand-labeled data, and then use that model to perform classification, for example, to identify the similarity between vectors or to classify vectors, and ultimately to identify the similarity between information or to classify information.

[0053] In addition, when the reference information and target information are images, the information processing device 1 can input information including the reference information, the target information, and instruction information indicating an instruction to output the similarity between the reference information and the target information to the multimodal generation AI as input information, and cause the multimodal generation AI to output the similarity.

[0054] In this case, the information processing device 1 classifies the impression IM into, for example, an impression IM corresponding to the reference information with the highest similarity generated by the multimodal generation AI, whose similarity is equal to or greater than a threshold, and an impression IM corresponding to the reference information with the highest similarity generated by the multimodal generation AI, whose similarity is equal to or greater than a threshold.

[0055] In addition, when the reference information and target information are images, the information processing device 1 can also have the multimodal generation AI generate an explanatory text that explains what images are included in each of the reference information and target information.

[0056] In this case, the information processing device 1 vectorizes each of the description of the reference information and the description of the target information, and calculates the similarity between the vectorized reference information and the vectorized target information. The method of calculating the similarity and the method of classifying the target information into impressions IM are the same as when the reference information and the target information are text.

[0057] When the reference information and the target information are images, the information processing device 1 can also directly vectorize each of the reference information and the target information to calculate the similarity.

[0058] Furthermore, when the reference information and the target information each contain text and an image, the information processing device 1 classifies the information into an impression IM corresponding to the reference information whose integrated score, which is a score obtained by weighting the similarity between the texts and the similarity between the images, is equal to or greater than a threshold, and an impression IM corresponding to the reference information whose integrated score is equal to or greater than the threshold and is the largest.

[0059] Next, the information processing device 1 determines the impression formed by the user U based on the user behavior information acquired in step S2 (step S4). If the improvement request received in step S1 includes determination method type information, the information processing device 1 determines the impression formed by the user U using the determination method of the type indicated by the determination method type information. If the improvement request received in step S1 does not include determination method type information, the information processing device 1 determines the impression formed by the user U using the first determination method.

[0060] First, the first determination method will be described. The information processing device 1 determines the impression that the user U has of a specific target from among a plurality of impressions IM1 to IMm included in the impression group using a generation AI.

[0061] The information processing device 1 uses a generation AI to determine the impression the user U has of the specific object from among the multiple impressions included in the impression group, based on the user behavior information acquired in step S2. First, the information processing device 1 causes the generation AI to determine the impression IM the user U has of the specific object from among the multiple impressions IM1 to IMm included in the impression group.

[0062] For example, the information processing device 1 inputs, as input information to the generation AI, information including instruction information instructing the user U to select an impression IM that the user U has had toward a specific target from among the multiple impressions IM1 to IMm indicated by the impression information, impression information indicating the multiple impressions IM1 to IMm included in the impression group, and user behavior information acquired in step S2, and causes the generation AI to output information indicating one or more impressions IM that are determined to have been held by the user U toward the specific target.

[0063] For example, suppose the impression group is an impression group based on the user U's values, the target information is a catchphrase "You can tell the difference because it's something you use every day," and the user behavior information is a posted message "Even beginners can drive it very easily!" In this case, the information processing device 1 includes, for example, a character string "Please select three or more of the values ​​below that represent the impression that the user who posted "Even beginners can drive it very easily!" has of a certain target.\nValues, Reason" in the input information as instruction information. In this way, the instruction information includes the user behavior information acquired in step S2.

[0064] Furthermore, the information processing device 1 includes information including the above-described impression information in the input information, similarly to the case of step S3.

[0065] Furthermore, the information processing device 1 can use, for example, information such as "Please select three or more of the values ​​below that represent the impression that you believe the user who posted this had," instead of the information in the string "Please determine a score for each of the values ​​below that indicates the degree of impression that you believe the user who posted this had. The score should be in the range of 1 to 10, and the higher the value, the higher the degree of impression.", but is not limited to such an example.

[0066] Furthermore, the information processing device 1 can also use a generation AI to determine the impression that the user U has of the target information from among multiple impressions IM1 to IMm included in the impression group, by limiting the attributes of the user U. In this case, instead of the information such as the string "Please select three or more of the values ​​below that are the impression that the user who posted this is judged to have had," the information processing device 1 can also use information such as "Please select three or more of the values ​​below that are the impression that the man in his twenties who posted this is judged to have had," or information such as the string "Please judge for each of the values ​​below a score indicating the level of impression that the man in his twenties who posted this had. The score should be in the range of 1 to 10, with a higher value indicating a higher level of impression." When the generation AI is to determine one or more impressions IM that a user U with an attribute other than a man in his twenties has had, the information of the string "man in his twenties" can be replaced with information indicating another attribute.

[0067] Furthermore, when the user behavior information is image information, the information processing device 1 can input information including target information, instruction information including information such as a string of characters "Please select three or more images from the values ​​below that fit the input image," and impression information as input information to the multimodal generation AI, and cause the multimodal generation AI to output a similarity. Note that the information processing device 1 can limit the attributes of the user U and output a score indicating the degree of impression IM, just as when the target information is other than image information.

[0068] Next, the second determination method will be described. The information processing device 1 causes the generation AI to generate reference information for determining that the user U has had the impression IM for each impression IM included in the impression group, and compares the reference information with the user behavior information to determine one or more impressions IM that the user U has had with respect to a specific target.

[0069] For each impression IM included in the impression group, the information processing device 1 generates information about the user U who has had that impression as reference information. When the impression group is an impression group based on values ​​based on Schwartz's value theory, the multiple impressions IM classified into the impression group are the 10 types of values ​​or 56 types of values ​​described above.

[0070] For example, the information processing device 1 inputs input information including instruction information, which is information instructing the generation AI to generate reference information that is determined to indicate that the user U has had the impression indicated by the impression IM, to the generation AI, and causes the generation AI to generate the reference information. The reference information is, for example, information indicated by at least one of text and an image.

[0071] The information processing device 1 stores standard instruction information in advance, and inputs information including the standard instruction information and information about the impression IM to the generation AI as instruction information, causing the generation AI to generate reference information for each impression IM. For example, if the specific object is a "car" and the user behavior information is information about reviews of the specific object, the standard instruction information is a character string such as "I will post a certain value. Please create 10 reviews about cars that Japanese users are likely to post that share that value. Please do not include too many of the posted values ​​directly in the text."

[0072] Moreover, the information on the impression IM is, for example, information on the character string “values: social power.” As a result, each of the 10 reviews that are predicted to give the user U the impression IM of social power is generated as reference information.

[0073] After generating the reference information for each impression IM, the information processing device 1 compares the user behavior information acquired in step S2 with the reference information for each impression IM. To compare the user behavior information with the reference information for each impression IM, the information processing device 1 vectorizes the user behavior information and each piece of reference information.

[0074] The vectorization of information is performed, for example, by embedding using a language model (for example, a transformer-based model). The vectorized information is represented, for example, by a vector with several hundred dimensions, but is not limited to such an example.

[0075] Embedding using a language model is, for example, embedding using text-embedding-ada provided by OpenAI, BERT, etc., but is not limited to these examples.

[0076] Note that the vectorization of information is not limited to embedding using a language model, and may be performed using, for example, Doc2Vec, the average of word embedding, etc. Word2Vec, fastText, etc. are used for word embedding.

[0077] The information processing device 1 classifies the user behavior information into an impression IM corresponding to reference information whose vector similarity with the user behavior information is equal to or greater than a threshold. The information processing device 1 can classify the user behavior information into only one impression IM, or into two or more impression IMs.

[0078] For example, the information processing device 1 can classify the user behavior information into only one impression IM by classifying the user behavior information into an impression IM corresponding to the reference information whose vector similarity is equal to or greater than a threshold and has the highest similarity.

[0079] Furthermore, when there are two or more pieces of reference information whose vector similarity with the user behavior information is equal to or greater than a threshold, the information processing device 1 can classify the user behavior information into two or more impressions IM corresponding to the two or more pieces of reference information, respectively.

[0080] The similarity between the vectors is cosine similarity, but may also be Jaccard similarity, or the inverse of the Euclidean distance or the inverse of the Manhattan distance. When the Euclidean distance or the Manhattan distance is used, the information processing device 1 classifies, for example, user behavior information whose Euclidean distance or Manhattan distance from the reference information is less than a threshold into an impression IM corresponding to the reference information.

[0081] In addition, the information processing device 1 can use any method to learn a classification model using vectors and the corresponding hand-labeled data, and then use that model to perform classification, for example, to identify the similarity between vectors or to classify vectors, and ultimately to identify the similarity between information or to classify information.

[0082] In addition, when the reference information and user behavior information are images, the information processing device 1 can input information including the reference information, the user behavior information, and instruction information indicating an instruction to output the similarity between the reference information and the user behavior information to the multimodal generation AI as input information, and cause the multimodal generation AI to output the similarity.

[0083] In this case, the information processing device 1 classifies the impression IM into, for example, an impression IM corresponding to the reference information with the highest similarity generated by the multimodal generation AI, whose similarity is equal to or greater than a threshold, and an impression IM corresponding to the reference information with the highest similarity generated by the multimodal generation AI, whose similarity is equal to or greater than a threshold.

[0084] In addition, when the reference information and user behavior information are images, the information processing device 1 can also have the multimodal generation AI generate an explanatory text that explains what images are included in each of the reference information and user behavior information.

[0085] In this case, the information processing device 1 vectorizes each of the descriptions of the reference information and the user behavior information, and calculates the similarity between the vectorized reference information and the vectorized user behavior information. The method of calculating the similarity and the impression group of the user behavior information to the impression IM are the same as when the reference information and the user behavior information are text.

[0086] When the reference information and the user behavior information are images, the information processing device 1 can also directly vectorize each of the reference information and the user behavior information to calculate the similarity.

[0087] Furthermore, when the reference information and the user behavior information each contain text and an image, the information processing device 1 classifies the user behavior information into impression IMs corresponding to the reference information whose integrated score, which is a score obtained by weighting and adding the similarity between the texts and the similarity between the images, is equal to or greater than a threshold, or impression IMs corresponding to the reference information whose integrated score is equal to or greater than the threshold and is the largest.

[0088] Next, the information processing device 1 merges the estimation result of step S3 and the determination result of step S4 (step S5). For example, as shown in Fig. 1, the information processing device 1 generates merged information including the estimation result of step S3 and the determination result of step S4. The merged information includes specific target information and user target information.

[0089] The specific target information includes information indicating the type of specific information, the target information included in the improvement request received in step S1, and the estimation result of step S3. In the example shown in Fig. 1, the information indicating the type of specific information includes information on the character string "text," the target information included in the improvement request received in step S1 includes information on the character string "You can tell the difference because it's something you use every day," and the estimation result of step S3 includes information on the character string "freedom, independence, self-respect."

[0090] The user target information also includes the type of user behavior information acquired in step S2, the user behavior information acquired in step S2, and the determination result of step S4. In the example shown in Fig. 1, the information indicating the type of user behavior information includes information on the character string "review", the user behavior information acquired in step S2 includes information on the character string "It's very easy to drive even for beginners!", and the determination result of step S4 includes information on the character string "freedom, fun".

[0091] Next, the information processing device 1 generates improvement information including information indicating improvements to the target information based on the merged information obtained by merging in step S5 (step S6). The information indicating improvements to the target information is, but is not limited to, the improved target information, information indicating a method for providing the improved target information, or information indicating a method for improving the target information. The improvement information includes one or more pieces of information indicating improvements to the target information.

[0092] The information processing device 1 generates a prompt, which is information to be input to the generation AI, based on information including, for example, supplemental information regarding the merge information, the merge information, and instruction information. For example, the information processing device 1 generates a prompt including supplemental information regarding the merge information, the merge information, and instruction information for causing the generation AI to output improvement information from the merge information. The information processing device 1 inputs the generated prompt to the generation AI and causes the generation AI to output the improvement information, thereby generating the improvement information.

[0093] The supplemental information includes information indicating the specific target, information indicating the relationship between the estimation result of step S3 and the determination result of step S4, etc. The supplemental information shown in Fig. 1 is the character string "There is the following gap between the advertising text when compact car model A is put on sale and the user's perception of compact car model A."

[0094] The instruction information includes a character string based on the improvement direction information included in the improvement request received in step S1. For example, if the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression that the user U is estimated to have, the instruction information includes, but is not limited to, the character string "Please consider delivery conditions for advertising targets to change the user's values."

[0095] Furthermore, if the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression determined to have been held by user U, the instruction information includes, but is not limited to, the string "Please change the advertisement text to something appropriate so that it matches the values ​​felt by the user."

[0096] The instruction information shown in Figure 1 is the string "There is such a gap, so please consider delivery conditions for the advertising target to change the user's values." Note that the information input to the generation AI may include information about the advertising target in addition to the information described above.

[0097] Next, the information processing device 1 provides the improvement information generated in step S6 to the worker O (step S7). For example, the information processing device 1 provides the improvement information generated in step S6 to the terminal device 3.

[0098] In this way, the information processing device 1 generates improvement information including information indicating improvements to be made to the target information, based on target information including information on the specific target, target impression information indicating the impression that the user U is estimated to have of the specific target, and user impression information indicating the impression that the user U is determined to have of the target information. This allows the information processing device 1 to improve the convenience of the provider of the specific target.

[0099] The configuration of an information processing system including an information processing device 1 that performs such processing, a plurality of terminal devices 2, a terminal device 3, and an information processing device 4 will be described in detail below.

[0100] [2. Information Processing System Configuration] 2 is a diagram illustrating an example of a configuration of an information processing system according to an embodiment. As illustrated in FIG. 2, the information processing system 100 according to the embodiment includes an information processing device 1, a plurality of terminal devices 2, a terminal device 3, and an information processing device 4.

[0101] The multiple terminal devices 2 are used by different users U, and the terminal device 3 is used by a worker O. The terminal devices 2 and 3 are, for example, notebook PCs (Personal Computers), desktop PCs, smartphones, tablet PCs, and wearable devices. Examples of wearable devices include, but are not limited to, smart glasses and smart watches.

[0102] The information processing device 4 provides various information to the user U via online sites. Examples of online sites provided by the information processing device 4 include, but are not limited to, e-commerce sites, news sites, Q&A sites, map sites, image posting sites, and video viewing sites.

[0103] The information processing device 1, the terminal device 2, the terminal device 3, and the information processing device 4 are connected to each other via a network N so as to be able to communicate with each other by wire or wirelessly. Note that the information processing system 100 shown in FIG. 2 may include a plurality of information processing devices 1, 4, etc.

[0104] 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: 5th generation mobile communication system).

[0105] The terminal devices 2, 3 and the information processing device 4 can connect to the network N and communicate with the information processing device 1 via short-range wireless communication such as a mobile communication network, Bluetooth (registered trademark), or wireless LAN (Local Area Network).

[0106] 3. Configuration of Information Processing Device 1 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.

[0107] [3.1. Communication Unit 10] The communication unit 10 is realized by, for example, a communication module or a network interface card (NIC). The communication unit 10 is connected to a 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.

[0108] [3.2. Storage section 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 has a user information storage unit 20.

[0109] 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" and "user information."

[0110] The "user ID" is identification information that identifies the user U. The "user information" is information about the user U that corresponds to the "user ID," and includes items such as "attribute information" and "behavioral history."

[0111] "Attribute information" is attribute information of user U corresponding to "user ID," and includes, for example, psychographic attribute information, demographic attribute information, etc. Demographic attributes include, for example, gender, age, place of residence, and occupation, while psychographic attributes include interests such as travel, clothing, cars, and religion, lifestyle, thoughts, and ideological tendencies.

[0112] "Behavior history" includes information on the behavior history of user U associated with "user ID." The behavior history of user U includes, for example, information on the movement history of user U and information on the behavior history of user U in online services. The information on the movement history of user U includes, for example, information on the route traveled by user U and information on places visited by user U.

[0113] The information on the behavior history of the user U in the online service includes information indicating the behavior of the user U regarding a specific target for which content is provided on an online site provided by the information processing device 4.

[0114] The user U's behavior regarding the specific target may include, for example, posting a review of the specific target, responding to a questionnaire about the specific target by the user U, evaluating the specific target posted by the user U (e.g., a positive evaluation, a negative evaluation, or the degree of the evaluation), browsing a web page of the specific target by the user U, searching for the specific target by the user U, etc. Furthermore, the behavior regarding the specific target may be a comment about the specific target on an SNS. A review of the specific target, a comment about the specific target on an SNS, and an evaluation of the specific target posted by the user U are examples of posted messages.

[0115] Furthermore, information on the user U's behavioral history in online services includes, for example, the user U's search history, access history, and posting history in online services. The user U's search history includes, for example, the user U's content search history in web search services. The user U's access history is, for example, the user U's content access history in online services, and includes information on the content of various sites that the user U accessed.

[0116] [3.3. Processing Unit 12] The processing unit 12 is a controller, and is realized by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs (corresponding to examples of information processing programs) stored in a storage device inside the information processing device 1 using RAM or the like as a working area.

[0117] The processing unit 12 is a controller, and may be realized by an integrated circuit such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or a general purpose graphic processing unit (GPGPU).

[0118] 3, the processing unit 12 has a receiving unit 30, an acquiring unit 31, an estimating unit 32, a determining unit 33, a generating unit 34, and a providing unit 35, and realizes or executes the functions and actions 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 may be any other configuration that performs the information processing described below.

[0119] 3.3.1. Reception unit 30 The reception unit 30 receives various information and requests from the terminal device 2 and the terminal device 3 via the network N and the communication unit 10.

[0120] For example, the receiving unit 30 receives an improvement request transmitted from the terminal device 3 of a worker O who creates or provides target information, which is information of a specific target. The improvement request includes, for example, target information including the information of the specific target, improvement direction information indicating a direction of improvement of the target information, and user identification information, which is information for identifying a target user U. By receiving the improvement request, the receiving unit 30 receives the target information, improvement direction information, and user identification information.

[0121] The specific target may be, for example, a product or a service, but may also be an organization such as a company, a facility such as a school or a hospital, a local public body such as a city, town, or village, or other target. The specific information may be an advertisement for promoting the specific target, such as a catchphrase, package, description, or introduction of the specific target, but is not limited to such examples. For example, if the specific target is video content or music content, the information about the specific target may be the video content or music content itself.

[0122] The direction of improvement of the target information is, for example, a direction of improvement toward the impression that user U is estimated to have, or a direction of improvement toward the impression that user U is determined to have had, but is not limited to such examples.

[0123] For example, the direction of improvement of the target information may be an intermediate direction of improvement between the impression that user U is estimated to have and the impression that user U is determined to have had, or it may be an improvement direction that indicates the degree to which the impression that user U is estimated to have had (e.g., closer to 70%), or it may be an improvement direction that indicates the degree to which the impression that user U is determined to have had (e.g., closer to 80%).

[0124] The user identification information is information for identifying a target user U. The target user U is a user U to whom the target information has been provided, but is not limited to such an example, and may be, for example, a user U to whom the target information has been provided and who has purchased or used a specific target. The target user U may also be a user U to whom the target information has been provided. The user identification information is, for example, information indicating the attributes of the target user U, information indicating the behavior of the target user U, which is the behavior of the user U regarding a specific target, and the like, but is not limited to such an example.

[0125] The user U's behavior regarding the specific target may be, for example, a review of the specific target or a response by the user U to a questionnaire about the specific target, but is not limited to such examples. The behavior regarding the specific target may also include an evaluation of the specific target posted by the user U (for example, a positive evaluation, a negative evaluation, or the degree of the evaluation), a browsing behavior of the web page of the specific target by the user U, a search behavior of the specific target by the user U, etc. The behavior regarding the specific target may also be a comment about the specific target on a social networking site.

[0126] The improvement request may also include estimation method type information indicating the type of estimation method for the impression the user U has of the specific target. There are multiple types of estimation methods for the impression the user U has of the specific target, including a first estimation method and a second estimation method. The receiving unit 30 receives the improvement request, thereby receiving the estimation method type information.

[0127] The first estimation method is a method in which the generation AI directly estimates the impression that user U has of a specific target. The second estimation method is a method in which the generation AI generates, for each impression included in the impression group, reference content that is estimated to have the impression that user U has, and the reference content is compared with the target content to estimate the impression that user U has of the target content.

[0128] The improvement request may also include determination method type information indicating the type of method for determining the impression the user U has of the specific target. There are multiple types of determination methods for the impression the user U has of the specific target, including a first determination method and a second determination method. The receiving unit 30 receives the determination method type information by receiving the improvement request.

[0129] The first determination method is a method in which the generation AI directly determines the impression that user U had of a specific target. The second estimation method is a method in which, for each impression included in the impression group, the generation AI generates reference content that is determined to be the impression that user U had, and compares the reference content with the target content to determine the impression that user U had of the target content.

[0130] [3.3.2. Acquisition part 31] The acquisition unit 31 acquires various pieces of information from the information processing device 4 and the storage unit 11. For example, the acquisition unit 31 acquires various pieces of content and information about each user U from the information processing device 4 via the network N and the communication unit 10. In addition, the acquisition unit 31 acquires information about each user U from the user information storage unit 20 of the storage unit 11.

[0131] The acquisition unit 31 includes a target-related information acquisition unit 40 that acquires target information including information about a specific target and target impression information indicating the impression that a user U is estimated to have of the specific target, and a user information acquisition unit 41 that acquires user impression information indicating the impression that the user U is determined to have of the target information. As described above, the specific target is, for example, a product or service. As described above, the specific information is an advertisement for promoting the specific target, such as a catchphrase, package, description, or introduction of the specific target, but is not limited to such examples.

[0132] 3.3.2.1. Object-related information acquisition unit 40 The object-related information acquiring unit 40 acquires the object information, the improvement directionality information, and the user identification information included in the improvement request received by the receiving unit 30 .

[0133] Furthermore, when the improvement request received by the receiving unit 30 includes estimation method type information, the object-related information acquiring unit 40 acquires the estimation method type information included in the improvement request.

[0134] Furthermore, the object-related information acquisition unit 40 acquires object impression information indicating the impression that the user U has of the specific object, which is estimated by the estimation unit 32, from the estimation unit 32. Furthermore, when the object impression information is stored in the storage unit 11 by the estimation unit 32, the object-related information acquisition unit 40 acquires the object impression information from the storage unit 11.

[0135] [3.3.2.2. User Information Acquisition Unit 41] The user information acquisition unit 41 acquires user behavior information from the information processing device 4 or the storage unit 11 based on the user identification information included in the improvement request received by the reception unit 30.

[0136] The user behavior information is information about the user U identified by the user identification information, and is information indicating the behavior of the user U regarding content provided on the online site provided by the information processing device 4. The content of the specific target provided on the online site provided by the information processing device 4 includes content for the user U to perform the behavior regarding the specific target described above, such as content including content for posting a review on the specific target, content including content for responding to a questionnaire about the specific target, etc., but is not limited to such examples.

[0137] The content of the specific target includes, but is not limited to, specific information, and may not include specific information. The target information may be, but is not limited to, an advertisement including one or more of a package of the specific target, a catchphrase of the specific target, a description of the specific target, and an introduction of the specific target. For example, if the specific target is content, the specific information may be information about the content itself.

[0138] As described above, the behavior of user U indicated by the user behavior information includes, for example, reviews of a specific target, responses by user U to a questionnaire about a specific target, evaluations of a specific target posted by user U (e.g., positive evaluations, negative evaluations, or the degree thereof), browsing behavior of user U on a web page of a specific target, search behavior of user U for a specific target, comments about a specific target on SNS, etc., but is not limited to such examples.

[0139] Furthermore, the user information acquisition unit 41 acquires user impression information indicating the impression that the user U has had of the target information, as determined by the determination unit 33, from the determination unit 33. Furthermore, when the user impression information is stored in the storage unit 11 by the determination unit 33, the user information acquisition unit 41 acquires the user impression information from the storage unit 11.

[0140] [3.3.3. Estimation section 32] The estimation unit 32 performs various estimations. The estimation unit 32 estimates the impression that the user U is likely to have, based on the object information. For example, the estimation unit 32 estimates the impression that the user U is likely to have, based on the object information acquired by the object-related information acquisition unit 40.

[0141] When the improvement request received by the receiving unit 30 includes estimation method type information, the estimation unit 32 estimates the impression held by the user U using an estimation method of the type indicated by the estimation method type information. When the improvement request received by the receiving unit 30 does not include estimation method type information, the information processing device 1 estimates the impression held by the user U using a first estimation method. The generation AI is arranged in an external information processing device, and the estimation unit 32 uses the generation AI via an API, but the generation AI may also be arranged within the information processing device 1.

[0142] First, the first estimation method will be described. The estimation unit 32 estimates the impression that the user U has of a specific target from among the multiple impressions IM1 to IMm included in the impression group, based on the target information included in the improvement request accepted by the accepting unit 30, using a generation AI.

[0143] The impressions IM1 to IMm are values ​​based on Schwartz's value theory, but are not limited to such examples. For example, the impressions IM1 to IMm may be values ​​classified as traditionalism, success orientation, self-actualization orientation, symbiosis orientation, etc., or other values. Furthermore, the impressions IM1 to IMm may be classified into five dimensions, such as sincerity, excitement, ability, refinement, and robustness, or further subdivided into these dimensions. They may also be impressions classified based on functional benefits such as performance, quality, and convenience, and emotional benefits such as joy, satisfaction, and security.

[0144] The impressions IM1 to IMm may be classified into four dimensions, such as quality, emotion, price, and social value, or further subdivided into these, or may be classified by other classification methods. The impressions IM1 to IMm may also be preset impressions for each type of specific object.

[0145] First, the estimation unit 32 causes the generation AI to estimate the impression IM that the user U has of the specific target from the multiple impressions IM1-IMm included in the impression group. For example, the information processing device 1 inputs, as input information to the generation AI, information including instruction information instructing the user to select the impression IM that the user U has of the specific target from the multiple impressions IM1-IMm indicated by the impression information, impression information indicating the multiple impressions IM1-IMm included in the impression group, and target information, and causes the generation AI to output information indicating one or more impressions IM that are estimated to be held by the user U of the specific target.

[0146] For example, suppose the impression group is an impression group based on the values ​​of user U, and the target information is a catchphrase, "You'll know the difference because it's something you use every day." In this case, the information processing device 1 includes, for example, a character string, "Please select three or more images from the values ​​below that fit the catchphrase, 'You'll know the difference because it's something you use every day.'\nValues, Reason," as instruction information in the input information. In this way, the instruction information includes the target information included in the improvement request accepted by the accepting unit 30.

[0147] In addition, the estimation unit 32 includes information including the character string "social power, authority, wealth, saving face, social recognition\n success, competence, ambition, influence, intelligence\n fun, enjoying life\n courage, a varied life, a vibrant life\n creativity, curiosity, freedom, choosing goals, self-respect, independence\n environmental protection, the world of beauty, harmony with nature, generosity, social justice, wisdom, equality, world peace, inner harmony\n assistance, honesty, tolerance, loyalty, responsibility, true friendship, the spiritual world, mature love, the meaning of life\n politeness, respect for parents and elders, self-discipline, obedience\n piety, acceptance of fate, humility, moderation, respect for tradition, detachment\n cleanliness, national security, social order, family security, giving back, health, a sense of belonging" in the input information as impression information.

[0148] Furthermore, instead of the information of the character string "Please select three or more images from the values ​​below that fit the catchphrase," the estimation unit 32 can also use information such as "Please select three or more impressions from the values ​​below that are estimated to be formed by the catchphrase," or information such as "Please estimate a score indicating the degree of impression that the user is estimated to have of the catchphrase for each of the values ​​below. The score should be in the range of 1 to 10, with a higher value being used for a higher degree of impression," but is not limited to such examples.

[0149] Furthermore, the estimation unit 32 can also use the generation AI to estimate the impression that the user U has of the target information from multiple impressions IM1 to IMm included in the impression group by limiting the attributes of the user U. In this case, instead of the information of the string "Please select three or more images from the values ​​below that fit the catchphrase," the estimation unit 32 can use information such as "Please select three or more impressions from the values ​​below that are estimated to be formed by a man in his twenties for the catchphrase." or information such as the string "Please estimate a score indicating the degree of impression that a man in his twenties is estimated to have for the catchphrase for each of the values ​​below. The score should range from 1 to 10, with a higher value indicating a higher degree of impression." When the generation AI is to estimate one or more impressions IM formed by a user U with an attribute other than a man in his twenties, the information of the string "man in his twenties" can be replaced with information indicating another attribute.

[0150] Furthermore, when the target information is image information, the estimation unit 32 can input information including the target information, instruction information including information such as a string of characters "Please select three or more images from the values ​​below that fit the input image," and impression information as input information to the multimodal generation AI, and cause the multimodal generation AI to output a similarity. Note that the information processing device 1 can limit the attributes of the user U and output a score indicating the degree of impression IM, just as when the target information is other than image information.

[0151] The estimation unit 32 can also use the generation AI to estimate the degree of impression that the user U is likely to have for each impression. For example, by inputting instruction information including information such as the character string "Please estimate the proportion of each value that the input catchphrase gives to the user" to the generation AI, the degree of impression that the user U is likely to have for each impression is estimated.

[0152] Next, the second estimation method will be described. The estimation unit 32 causes the generation AI to generate reference information that estimates that the user U will have the impression IM for each impression IM included in the impression group, and compares the reference information with the target information to estimate one or more impressions IM that the user U will have of the target information.

[0153] For each impression IM included in the impression group, the estimation unit 32 generates reference information that estimates that the user U will have that impression. When the impression group is an impression group based on values ​​based on Schwartz's value theory, the multiple impressions IM classified into the impression group are the 10 types of values ​​or 56 types of values ​​described above.

[0154] For example, the estimation unit 32 inputs input information including instruction information, which is information instructing the generation AI to generate reference information that is estimated to have the impression indicated by the impression IM held by the user U, to the generation AI, and causes the generation AI to generate the reference information. The reference information is, for example, information indicated by at least one of text and an image.

[0155] The estimation unit 32 stores standard instruction information in advance, and inputs information including the standard instruction information and information about the impression IM to the generation AI as instruction information, causing the generation AI to generate reference information for each impression IM. For example, if the specific object is a "car" and the target information is a catchphrase for the specific object, the standard instruction information is the character string "I will post a certain value. Please create 10 catchphrases for cars that are likely to share the value that Japanese users have. Please do not include too many of the posted values ​​directly in the sentences."

[0156] Moreover, the information on the impression IM is, for example, information on the character string “values: social power.” As a result, each of the 10 catchy slogans that are predicted to give the user U the impression IM of social power is generated as reference information.

[0157] After generating the reference information for each impression IM, the estimation unit 32 compares the target information that is the target of estimation of the impression IM held by the user U with the reference information for each impression IM. In order to compare the target information with the reference information for each impression IM, the estimation unit 32 vectorizes the target information and each piece of reference information.

[0158] The vectorization of information is performed, for example, by embedding using a language model (for example, a transformer-based model). The vectorized information is represented, for example, by a vector with several hundred dimensions, but is not limited to such an example.

[0159] Embedding using a language model is, for example, embedding using text-embedding-ada provided by OpenAI, BERT, etc., but is not limited to these examples.

[0160] Note that the vectorization of information is not limited to embedding using a language model, and may be performed using, for example, Doc2Vec, the average of word embedding, etc. Word2Vec, fastText, etc. are used for word embedding.

[0161] The estimation unit 32 classifies the target information into an impression IM corresponding to reference information whose vector similarity with the target information is equal to or greater than a threshold value. The estimation unit 32 can classify the target information into only one impression IM, or into two or more impression IMs.

[0162] For example, the estimation unit 32 can classify the target information into only one impression IM by classifying the target information into an impression IM corresponding to the reference information whose vector similarity is equal to or greater than a threshold and has the highest similarity.

[0163] Furthermore, when there are two or more pieces of reference information whose vector similarity with the target information is equal to or greater than a threshold, the estimation unit 32 can classify the target information into two or more impressions IM corresponding to the two or more pieces of reference information, respectively.

[0164] The similarity between the vectors is cosine similarity, but may also be Jaccard similarity, or the inverse of the Euclidean distance or the inverse of the Manhattan distance. When the Euclidean distance or the Manhattan distance is used, the estimation unit 32 classifies, for example, object information whose Euclidean distance or the Manhattan distance from the reference information is less than a threshold into an impression IM corresponding to the reference information.

[0165] In addition, the estimation unit 32 can use any method to learn a classification model using vectors and the corresponding hand-labeled data, and then use that model to perform classification. This can also be used to identify the similarity between vectors, classify vectors, and ultimately identify the similarity between information and classify information.

[0166] In addition, when the reference information and target information are images, the estimation unit 32 can input information including the reference information, the target information, and instruction information indicating an instruction to output the similarity between the reference information and the target information to the multimodal generation AI as input information, and cause the multimodal generation AI to output the similarity.

[0167] In this case, the estimation unit 32 classifies the impression IM into, for example, an impression IM generated by a multimodal generation AI that corresponds to the reference information with the highest similarity, which is equal to or greater than a threshold, and an impression IM generated by a multimodal generation AI that corresponds to the reference information with a similarity equal to or greater than a threshold.

[0168] In addition, when the reference information and target information are images, the estimation unit 32 can also have the multimodal generation AI generate an explanatory text that explains what images are contained in each of the reference information and target information.

[0169] In this case, the estimation unit 32 vectorizes each of the description of the reference information and the description of the target information, and calculates the similarity between the vectorized reference information and the vectorized target information. The method of calculating the similarity and the method of classifying the target information into impressions IM are the same as when the reference information and the target information are text.

[0170] When the reference information and the target information are images, the estimation unit 32 can also directly vectorize each of the reference information and the target information to calculate the similarity.

[0171] Furthermore, when the reference information and the target information each contain text and an image, the estimation unit 32 classifies the reference information into an impression IM corresponding to the reference information whose integrated score, which is a score obtained by weighting and adding the similarity between the texts and the similarity between the images, is equal to or greater than a threshold, or an impression IM corresponding to the reference information whose integrated score is equal to or greater than the threshold and is the largest.

[0172] The estimation unit 32 can also estimate the above-mentioned similarity or overall score as the degree of impression that the user U is estimated to have.

[0173] [3.3.4. Judgment unit 33] The determination unit 33 determines the impression that the user U has based on the target information and the information of the user U regarding the specific target. For example, the determination unit 33 determines the impression that the user U is estimated to have had based on the user behavior information acquired by the user information acquisition unit 41.

[0174] As described above, user behavior information is information about user U identified by user identification information, and is information that indicates the behavior of user U regarding a specific target for which content (e.g., content such as an introduction page or purchase page for the specific target) is provided on an online site provided by information processing device 4.

[0175] The user behavior information is, for example, a message posted by the user U regarding the specific target, and the determination unit 33 determines the impression that the user U has based on the message posted by the user U regarding the specific target. The posted message is, for example, a review of the specific target, a comment on the specific target on SNS, an evaluation of the specific target posted by the user U, etc., but is not limited to these examples.

[0176] When the improvement request received by the receiving unit 30 includes determination method type information, the determination unit 33 determines the impression formed by the user U using the determination method of the type indicated by the determination method type information. When the improvement request received by the receiving unit 30 does not include determination method type information, the determination unit 33 determines the impression formed by the user U using the first determination method.

[0177] First, the first determination method will be described. The determination unit 33 determines the impression that the user U has of a specific target from among the multiple impressions IM1 to IMm included in the impression group using a generation AI.

[0178] The determination unit 33 determines the impression the user U has of the specific object from among the multiple impressions included in the impression group using the generation AI based on the user behavior information acquired in step S2. First, the determination unit 33 causes the generation AI to determine the impression IM the user U has of the specific object from among the multiple impressions IM1 to IMm included in the impression group.

[0179] For example, the determination unit 33 inputs, as input information to the generation AI, information including instruction information instructing the user U to select an impression IM that the user U has had toward a specific target from among the multiple impressions IM1 to IMm indicated by the impression information, impression information indicating the multiple impressions IM1 to IMm included in the impression group, and the user behavior information acquired in step S2, and causes the generation AI to output information indicating one or more impressions IM that the user U is estimated to have had toward the specific target.

[0180] For example, suppose the impression group is an impression group based on the user U's values, the target information is a catchphrase "You can tell the difference because it's something you use every day," and the user behavior information is a review (an example of a posted message) that reads "It's very easy to drive, even for beginners!" In this case, the determination unit 33 includes, as instruction information, information such as a character string "Please select three or more of the values ​​below that represent the impression that the user who posted "It's very easy to drive, even for beginners!" has of the target.\nValues, Reason" in the input information. In this way, the instruction information includes the user behavior information acquired in step S2.

[0181] Furthermore, similar to the case of the estimation process by the estimation unit 32, the determination unit 33 includes information including the above-described impression information in the input information.

[0182] Furthermore, the determination unit 33 can use, for example, information such as "Please select three or more of the values ​​below that represent the impression that you believe the user who posted this had," instead of the information such as "Please determine a score for each of the values ​​below that indicates the degree of impression that you believe the user who posted this had. The score should be in the range of 1 to 10, and the higher the value, the higher the degree of impression.", but is not limited to such an example.

[0183] Furthermore, the determination unit 33 can limit the attributes of the user U and use the generation AI to determine the impression the user U has of the target information from among the multiple impressions IM1 to IMm included in the impression group. In this case, instead of the information such as the string "Please select three or more of the values ​​below to indicate the impression that the user who posted the following is determined to have had," the determination unit 33 can use information such as "Please select three or more of the values ​​below to indicate the impression that the man in his twenties who posted the following is determined to have had," or information such as the string "Please determine a score indicating the level of impression that the man in his twenties who posted the following has had for each of the values ​​below. The score should be in the range of 1 to 10, with a higher value indicating a higher level of impression." When the generation AI is to determine one or more impressions IM that a user U with an attribute other than a man in his twenties has had, the information of the string "man in his twenties" can be replaced with information indicating another attribute.

[0184] Furthermore, when the user behavior information is image information, the determination unit 33 can input information including target information, instruction information including information such as a string of characters "Please select three or more images from the values ​​below that fit the input image," and impression information as input information to the multimodal generation AI, and cause the multimodal generation AI to output a similarity. Note that, as in the case where the target information is other than image information, the determination unit 33 can limit the attributes of the user U and output a score indicating the degree of impression IM.

[0185] Furthermore, the determination unit 33 can also use the generation AI to determine the degree of the impression that is determined to have been held by the user U for each impression. For example, by inputting instruction information including information such as the character string "Please determine the proportion of each value that the input catchphrase gives to the user" to the generation AI, the degree of the impression that is determined to have been held by the user U is determined for each impression.

[0186] Next, the second determination method will be described. The determination unit 33 causes the generation AI to generate reference information for determining that the user U has had the impression IM for each impression IM included in the impression group, and compares the reference information with the user behavior information to determine one or more impressions IM that the user U has had with respect to a specific target.

[0187] The determination unit 33 generates, for each impression IM included in the impression group, information about the user U who has had that impression as reference information. When the impression group is an impression group based on values ​​based on Schwartz's value theory, the multiple impressions IM classified into the impression group are the 10 types of values ​​or 56 types of values ​​described above.

[0188] For example, the determination unit 33 inputs input information including instruction information, which is information instructing the generation AI to generate reference information that is determined to indicate that the user U has had the impression indicated by the impression IM, to the generation AI, and causes the generation AI to generate the reference information. The reference information is, for example, information indicated by at least one of text and an image.

[0189] The determination unit 33 stores standard instruction information in advance, and inputs information including the standard instruction information and information about the impression IM to the generation AI as instruction information, thereby causing the generation AI to generate reference information for each impression IM. For example, if the specific object is a "car" and the user behavior information is information about reviews of the specific object, the standard instruction information is a character string such as "I will post a certain value. Please create 10 reviews about cars that Japanese users are likely to post that share that value. Please do not include too many of the posted values ​​directly in the text."

[0190] Moreover, the information on the impression IM is, for example, information on the character string “values: social power.” As a result, each of the 10 reviews that are predicted to give the user U the impression IM of social power is generated as reference information.

[0191] After generating the reference information for each impression IM, the determination unit 33 compares the user behavior information acquired by the acquisition unit 31 with the reference information for each impression IM. To compare the user behavior information with the reference information for each impression IM, the determination unit 33 vectorizes the user behavior information and each piece of reference information.

[0192] The vectorization of information is performed, for example, by embedding using a language model (for example, a transformer-based model). The vectorized information is represented, for example, by a vector with several hundred dimensions, but is not limited to such an example.

[0193] Embedding using a language model is, for example, embedding using text-embedding-ada provided by OpenAI, BERT, etc., but is not limited to these examples.

[0194] Note that the vectorization of information is not limited to embedding using a language model, and may be performed using, for example, Doc2Vec, the average of word embedding, etc. Word2Vec, fastText, etc. are used for word embedding.

[0195] The determination unit 33 classifies the user behavior information into an impression IM corresponding to reference information whose vector similarity with the user behavior information is equal to or greater than a threshold. The determination unit 33 can classify the user behavior information into only one impression IM, or into two or more impression IMs.

[0196] For example, the determination unit 33 can classify the user behavior information into only one impression IM by classifying the user behavior information into an impression IM corresponding to the reference information whose vector similarity is equal to or greater than a threshold and has the highest similarity.

[0197] Furthermore, when there are two or more pieces of reference information whose vector similarity with the user behavior information is equal to or greater than a threshold, the determination unit 33 can classify the user behavior information into two or more impressions IM corresponding to the two or more pieces of reference information, respectively.

[0198] The similarity between the vectors is cosine similarity, but may also be Jaccard similarity, or the inverse of the Euclidean distance or the inverse of the Manhattan distance. When the Euclidean distance or the Manhattan distance is used, the determination unit 33 classifies, for example, user behavior information whose Euclidean distance or Manhattan distance from the reference information is less than a threshold value into an impression IM corresponding to the reference information.

[0199] In addition, the judgment unit 33 can use any method to learn a classification model using vectors and the corresponding hand-labeled data, and then use that model to perform classification. This can also be used to determine the similarity between vectors, classify vectors, and ultimately identify the similarity between information and classify information.

[0200] In addition, when the reference information and user behavior information are images, the judgment unit 33 can input information including the reference information, the user behavior information, and instruction information indicating an instruction to output the similarity between the reference information and the user behavior information to the multimodal generation AI as input information, and cause the multimodal generation AI to output the similarity.

[0201] In this case, the determination unit 33 classifies the impression IM into, for example, an impression IM that corresponds to the reference information generated by the multimodal generation AI whose similarity is equal to or greater than a threshold and has the highest similarity, or an impression IM that corresponds to the reference information generated by the multimodal generation AI whose similarity is equal to or greater than a threshold.

[0202] In addition, when the reference information and user behavior information are images, the judgment unit 33 can also cause the multimodal generation AI to generate an explanatory text that explains what images are included in each of the reference information and user behavior information.

[0203] In this case, the determination unit 33 vectorizes each of the descriptions of the reference information and the user behavior information, and calculates the similarity between the vectorized reference information and the vectorized user behavior information. The method of calculating the similarity and the impression group of the user behavior information to the impression IM are the same as when the reference information and the user behavior information are text.

[0204] When the reference information and the user behavior information are images, the determination unit 33 can also directly vectorize each of the reference information and the user behavior information to calculate the similarity.

[0205] Furthermore, when the reference information and the user behavior information each contain text and an image, the determination unit 33 classifies the user behavior information into an impression IM corresponding to the reference information whose integrated score, which is a score obtained by weighting and adding the similarity between the texts and the similarity between the images, is equal to or greater than a threshold, or an impression IM corresponding to the reference information whose integrated score is equal to or greater than the threshold and is the largest.

[0206] The determination unit 33 can also determine the degree of impression that the user U is determined to have had using the above-mentioned similarity or overall score.

[0207] [3.3.5. Generation unit 34] The generation unit 34 generates improvement information including information indicating improvements regarding the target information, based on the target information and target impression information acquired by the target-related information acquisition unit 40 and the user impression information acquired by the user information acquisition unit 41. The generation unit 34 generates the improvement information using, for example, a generation AI.

[0208] As described above, the target information is information about a specific target. As described above, the target impression information is information indicating the impression that the user U has about the specific target, which is estimated by the estimation unit 32. The user impression information is information indicating the impression that the user U has about the target information, which is determined by the determination unit 33.

[0209] Furthermore, the generating unit 34 can generate improvement information based on information that further includes improvement directionality information received by the receiving unit 30 in addition to the target information, target impression information, and user impression information.

[0210] The generation unit 34 generates merged information by merging, for example, target information, target impression information, and user impression information, and generates a prompt to be input to the generation AI based on the generated merged information.The generation unit 34 then inputs the generated prompt to the generation AI as input information, causing the generation AI to output improvement information.The processing of the generation unit 34 will be specifically described below.

[0211] The generation unit 34 merges the estimation result by the estimation unit 32 and the determination result by the determination unit 33. The merged information includes, for example, specific target information and user target information.

[0212] The specific target information includes information indicating the type of specific information, target information included in the improvement request accepted by the accepting unit 30, and target impression information which is the estimation result by the estimating unit 32. The user target information includes the type of user behavior information acquired by the acquiring unit 31, the user behavior information acquired by the acquiring unit 31, and user target information which is the determination result by the determining unit 33.

[0213] 5 is a diagram showing an example of merge information generated by the generation unit 34 in the processing unit 12 of the information processing device 1 according to the embodiment. The merge information 50 shown in FIG. 5 includes information on the character string "text" as information indicating the type of specific information, information on the character string "you can tell the difference because it's something you use every day" as target information included in the improvement request accepted by the acceptance unit 30, and information on the character string "freedom, independence, self-respect" as an estimation result by the estimation unit 32.

[0214] In addition, the merge information 50 includes the information of the character string "review" as information indicating the type of user behavior information, the information of the character string "It's very easy to drive even for beginners!" as user behavior information acquired by the acquisition unit 31, and the information of the character string "freedom, fun" as the judgment result by the judgment unit 33.

[0215] The generation unit 34 generates improvement information including information indicating improvements to the target information based on the generated merge information. The information indicating improvements to the target information is, but is not limited to, the improved target information, information indicating a method for providing the improved target information, or information indicating a method for improving the target information. The improvement information includes one or more pieces of information indicating improvements to the target information.

[0216] The improved target information is, for example, an improved text advertisement if the target information is a text advertisement, an improved image advertisement if the target information is an image advertisement, an improved advertisement including at least one of text and image if the target information is an advertisement including text and image. The improved target information is, for example, an improved video content if the target information is video content, and an improved music content if the target information is video content.

[0217] For example, if the target information is an advertisement, the method of providing the target information may be, but is not limited to, sponsored search keywords, delivery conditions to the advertisement target, conditions of the target user, site traffic, etc. For example, if the target information is video content or music content, the method of providing the target information may be attribute conditions of the user to be recommended.

[0218] For example, if the target information is a text advertisement, the information indicating a method for improving the target information is information indicating the font, size, highlighting, placement, etc. of the text, and if the target information is an image advertisement, the information indicating a size, placement, etc. of the image. Furthermore, if the target information is an advertisement including text and an image, the information indicating a method for improving the target information is information indicating the font, size, highlighting, placement, etc. of the text, and the size, placement, etc. of the image.

[0219] Furthermore, when the target information is video content, the information indicating how to improve the target information is information indicating the improved scenario, script, cast, etc. When the target information is music content, the information indicating how to improve the target information is information indicating how to improve lyrics, information indicating the arrangement and selection of music components after improvement, etc., but is not limited to these examples.

[0220] The generation unit 34 generates a prompt, which is information to be input to the generation AI, based on information including, for example, supplemental information regarding the merge information, the merge information, and instruction information. For example, the generation unit 34 generates a prompt including supplemental information regarding the merge information, the merge information, and instruction information for causing the generation AI to output improvement information from the merge information. The generation unit 34 inputs the generated prompt to the generation AI and causes the generation AI to output the improvement information, thereby generating the improvement information.

[0221] The supplemental information includes information indicating the specific target, information indicating the relationship between the estimation result by the estimation unit 32 and the determination result by the determination unit 33, and the like.

[0222] The instruction information includes a character string based on the improvement direction information included in the improvement request received by the receiving unit 30. For example, if the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression that the user U is estimated to have, the instruction information includes, but is not limited to, the character string "Please consider delivery conditions for advertising targets to change the user's values."

[0223] Furthermore, if the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression determined to have been held by user U, the instruction information includes, but is not limited to, the string "Please change the advertisement text to something appropriate so that it matches the values ​​felt by the user."

[0224] The generation unit 34 has template information, which is information that serves as a template for each improvement direction indicated in the improvement direction information, and can generate a prompt using the template information that corresponds to the improvement direction indicated in the improvement direction information.

[0225] Fig. 6 is a diagram showing an example of template information used by the generating unit 34 in the processing unit 12 of the information processing device 1 according to the embodiment. Fig. 7 is a diagram showing an example of a prompt generated by the generating unit 34 in the processing unit 12 of the information processing device 1 according to the embodiment using the template information shown in Fig. 6.

[0226] When the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression determined to be held by the user U, the generation unit 34 generates a prompt using template information 60 shown in FIG.

[0227] Template information 60 shown in FIG. 6 is the following information: "There is a gap between the (type of advertisement) in the advertisement when selling (product) and the user's values ​​regarding (product).\n\n(merged content)\n\nBecause there is such a gap, please change the (type of advertisement) in the advertisement to an appropriate one that matches the user's values ​​regarding (product)."

[0228] 6, "(product)" is information indicating a specific target. When the specific target is a product, "(product)" is information indicating the product name, etc., and when the specific target is a service, "(product)" is information indicating the service name, etc.

[0229] 6, "(Type of advertisement)" is information indicating the type of target information. If the type of target information is a character string, "(Type of advertisement)" is "text," and if the type of target information is an image, "(Type of advertisement)" is "image." Also, in FIG. 6, "(Merged content)" is merge information generated by the generation unit 34.

[0230] When the "(product)" is "compact car model A", the "(type of advertisement)" is "text", and the "(merged content)" is the merge information shown in FIG. 5, the generation unit 34 generates, for example, the prompt 70 shown in FIG. 7.

[0231] When the improvement direction indicated by the improvement direction information is a direction of improvement toward the impression that the user U is estimated to have, the generation unit 34 generates a prompt using the template information 61 shown in FIG.

[0232] Fig. 8 is a diagram showing another example of template information used by the generating unit 34 in the processing unit 12 of the information processing device 1 according to the embodiment. Fig. 9 is a diagram showing an example of a prompt generated by the generating unit 34 in the processing unit 12 of the information processing device 1 according to the embodiment using the template information shown in Fig. 8.

[0233] Template information 61 shown in Figure 8 is the following information: "There is a gap between the (type of advertisement) used in advertisements when selling (product) and the user's perception of (product).\n\n(Merged content)\n\nBecause such a gap exists, please consider (ways to fill the gap*) to change the user's perception of (product).\n\n*Ways to fill the gap can be one or more of the following:\n·Consider the keywords for sponsored searches.\n·Consider the conditions for delivering advertisements to the target audience.\n·Create an image of the target users.\n·Consider the site flow, etc."

[0234] 8, "(product)" is information indicating a specific target. When the specific target is a product, "(product)" is information indicating the product name, etc., and when the specific target is a service, "(product)" is information indicating the service name, etc.

[0235] Also, in FIG. 8, "(type of advertisement)" is information indicating the type of target information. If the type of target information is a character string, "(type of advertisement)" is information of the character string "text," and if the type of target information is an image, "(type of advertisement)" is information of the character string "image." Also, in FIG. 8, "(merged content)" is merge information generated by the generation unit 34.

[0236] When the "(product)" is "compact car model A", the "(type of advertisement)" is "text", and the "(merged content)" is the merge information shown in FIG. 5, the generation unit 34 generates, for example, the prompt 71 shown in FIG. 9.

[0237] In prompt 71 shown in Figure 9, "Consider the conditions for delivery to the advertising target" is selected as a way to fill the gap from among "Consider the keywords for the sponsored search," "Consider the conditions for delivery to the advertising target," "Create an image of the target user," and "Consider the site traffic flow, etc."

[0238] The generation unit 34 inputs information including, for example, information indicating a specific target, information indicating the type of target information, merge information, template information 61, and instruction information to the generation AI as input information, and can cause the generation AI to output a prompt. In this case, the instruction information is, for example, information that uses template information 61 to instruct the generation AI to output a prompt from the information indicating a specific target, information indicating the type of target information, and merge information.

[0239] In addition, the generation unit 34 has information linking one or more types of gap filling methods with the type of specific target, and can also generate prompts on a rule-based basis based on information indicating one or more gap filling methods according to the type of specific target, information indicating the specific target, information indicating the type of target information, merge information, and template information 61.

[0240] The generation unit 34 can input, for example, a prompt 70 shown in Fig. 7 to the generation AI and cause the generation AI to output improvement information including one or more pieces of improved target information. Also, the generation unit 34 can input, for example, a prompt 71 shown in Fig. 9 to the generation AI and cause the generation AI to output improvement information including information indicating one or more improved methods of providing the target information.

[0241] 6 is information used to generate a prompt for outputting the improved target information, but may also be information used to generate a prompt for outputting information indicating a method for providing the improved information or information indicating a method for improving the target information. Similarly, the template information 61 shown in Fig. 8 is information used to generate a prompt for outputting information indicating a method for providing the improved information, but may also be information used to generate a prompt for outputting the improved target information or information indicating a method for improving the target information.

[0242] In addition, the generation unit 34 also has template information for each direction of improvement of the target information, such as template information for a direction of improvement that is intermediate between the impression that user U is estimated to have and the impression that user U is determined to have had, template information for a direction of improvement that indicates the degree to which the impression that user U is estimated to have (for example, closer to 70%), and template information for a direction of improvement that indicates the degree to which the impression that user U is determined to have had (for example, closer to 80%).

[0243] The template information for the direction of improvement indicating the degree of bias toward the impression that user U is estimated to have (for example, toward 70%) is, for example, information such as the character string "Consider (how to fill the gap*) to change the user's values ​​so that the (type of advertisement) when selling (product) is closer to the user's values ​​toward (product) by (percentage) compared to the user's values ​​toward (product)," but is not limited to such an example. In the case of being closer to 70%, the "(percentage)" is 80%.

[0244] Furthermore, the template information for the direction of improvement indicating the degree of proximity to the impression determined to have been held by user U (for example, closer to 80%) is, for example, information such as a character string "When selling (product), please change the (type of advertisement) of the advertisement to an appropriate one so that it is closer to the user's values ​​toward (product) by about (percentage) compared to the (type of advertisement) of the advertisement and the user's values ​​toward (product)." However, it is not limited to such an example. In the case of closer to 80%, the "(percentage)" is 70%.

[0245] Furthermore, when the estimation unit 32 estimates the degree of the impression that the user U is estimated to have for each impression, the generation unit 34 can include, in the merge information, information indicating the degree of each impression estimated by the estimation unit 32. Similarly, when the determination unit 33 estimates the degree of the impression that the user U is determined to have had for each impression, the generation unit 34 can include, in the merge information, information indicating the degree of each impression determined by the determination unit 33.

[0246] The generation unit 34 can also generate, for example, pre- and post-improvement impression information, which is information (e.g., a radar chart, etc.) that indicates the change in the degree of each impression that the user U has before and after improvement based on the improvement information, as estimated by the estimation unit 32.

[0247] In addition, the generation unit 34 can also generate, as the before-and-after improvement impression information, information (e.g., a radar chart) that indicates the change in the degree of each impression that the user U had before and after the improvement based on the improvement information, as determined by the determination unit 33.

[0248] The generation unit 34 can also generate impression change information (e.g., radar chart, line graph, bar graph, etc.) as time-series information, which indicates the change over time in the degree of each impression estimated by the estimation unit 32 that the user U has.

[0249] The generation unit 34 can also generate impression change information (e.g., radar chart, line graph, bar graph, etc.) as time-series information, which indicates the change over time in the degree of each impression determined by the determination unit 33 to have been held by the user U.

[0250] In addition, the generation unit 34 can also generate, as impression deviation information, information (e.g., a radar chart) that shows a comparison between the degree of each impression estimated by the estimation unit 32 that the user U has and the degree of each impression determined by the determination unit 33 that the user U has.

[0251] [3.3.6.Providing Department 35] The providing unit 35 provides various pieces of information. For example, the providing unit 35 provides the worker O with various pieces of information by transmitting the various pieces of information to the terminal device 3.

[0252] The providing unit 35 provides, for example, the improvement information generated by the generating unit 34 to the worker O. For example, the providing unit 35 provides the improvement information generated by the generating unit 34 to the worker O by transmitting the improvement information generated by the generating unit 34 to the terminal device 3.

[0253] In addition, the providing unit 35 can also provide the information selected by the worker O from the before-and-after-improvement impression information, impression change information, and impression deviation information generated by the generating unit 34 to the terminal device 3, by transmitting the information selected by the worker O from the before-and-after-improvement impression information, impression change information, and impression deviation information generated by the generating unit 34 to the worker O.

[0254] 10 is a diagram showing an example of the before-and-after impression information provided by the providing unit 35 in the processing unit 12 of the information processing device 1 according to the embodiment. In the example shown in FIG. 10, for ease of explanation, the impressions IM1 to IMm are classified into four categories: healthy, fun, joy, and excitement / thrill, and the impression levels are set to 0.1 to -0.02, but the present invention is not limited to such an example.

[0255] The before-and-after-improvement impression information 75 shown in Fig. 10 is a radar chart showing the level of each impression determined by the determination unit 33 as having been held by the user U before and after improvement based on the improvement information. In the example shown in Fig. 10, the impressions determined as having been held by the user U were high in excitement and nervousness before the improvement based on the improvement information, but high in enjoyment and health after the improvement based on the improvement information. In this way, the impression held by the user U can be changed by the improvement information generated by the information processing device 1.

[0256] [4. Processing Procedure] Next, a procedure of information processing by the processing unit 12 of the information processing device 1 according to the embodiment will be described. Fig. 11 is a flowchart showing an example of information processing by the processing unit 12 of the information processing device 1 according to the embodiment.

[0257] 11, the processing unit 12 of the information processing device 1 determines whether or not an improvement request has been received (step S10). If it is determined that an improvement request has been received (step S10: Yes), the processing unit 12 estimates the impression that the user U has, using a generation AI, based on the target information included in the improvement request (step S11).

[0258] Next, the processing unit 12 determines the impression that the user U has, using the generation AI, based on the user behavior information corresponding to the user identification information included in the improvement request (step S12).

[0259] Next, the processing unit 12 generates merged information by merging the estimation result in step S11 and the determination result in step S12 (step S13), and then generates a prompt including the merged information generated in step S13 (step S14).

[0260] Next, the processing unit 12 inputs the prompt generated in step S14 to the generation AI and causes the generation AI to output improvement information, thereby generating improvement information (step S15).Then, the processing unit 12 provides the improvement information generated in step S15 (step S16).

[0261] When the processing of step S16 is completed or when it is determined that an improvement request has not been received (step S10: No), the processing unit 12 determines whether or not it is time to end the operation (step S17). The processing unit 12 determines that it is time to end the operation when, for example, the power of the information processing device 1 is turned off.

[0262] If the processing unit 12 determines that the operation end time has not yet arrived (step S17: No), it proceeds to step S10, and if it determines that the operation end time has arrived (step S17: Yes), it terminates the processing shown in Figure 11.

[0263] [5. Modifications] In the above example, the generation unit 34 generates a prompt using a generation AI, but it is also possible to have a prompt for each combination of the type of identified target, the type of target information, and the improvement direction indicated by the improvement direction information, and to select a prompt according to the combination of the type of identified target, the type of target information, and the improvement direction indicated by the improvement direction information. In this case, the generation unit 34 can also cause the generation AI to output improvement information using the selected prompt.

[0264] Furthermore, the generation unit 34 can also generate or select template information according to the time-series changes in each impression determined by the determination unit 33, and generate a prompt using the generated or selected template information.

[0265] The impression groups used by the estimation unit 32 and the determination unit 33 are predetermined impression groups, but may also be impression groups selected by the worker O. Furthermore, the estimation unit 32 and the determination unit 33 have information on impression groups that differ for each behavior or attribute of the user U indicated in the user identification information, and can use impression groups according to the behavior or attribute of the user U indicated in the user identification information. Furthermore, the estimation unit 32 and the determination unit 33 have information on impression groups that differ for each type of specific target, and can use impression groups according to the type of specific target.

[0266] [6. Hardware Configuration] The information processing device 1 according to the embodiment described above is realized by, for example, a computer 80 configured as shown in Fig. 12. Fig. 12 is a hardware configuration diagram showing an example of the computer 80 that realizes the functions of the information processing device 1 according to the embodiment. The computer 80 has 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.

[0267] The CPU 81 operates and controls each part based on programs stored in the ROM 83 or the HDD 84. The ROM 83 stores a boot program executed by the CPU 81 when the computer 80 starts up, programs that depend on the hardware of the computer 80, and the like.

[0268] The HDD 84 stores programs executed by the CPU 81, data used by such programs, etc. The communication interface 85 receives data from other devices via the network N (see FIG. 2) and sends it to the CPU 81, and transmits data generated by the CPU 81 to other devices via the network N.

[0269] 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 devices via the input / output interface 86. The CPU 81 also outputs generated data to the output devices via the input / output interface 86.

[0270] The media interface 87 reads a program or data stored in a recording medium 88 and provides it to the CPU 81 via the RAM 82. The CPU 81 loads the program or data 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 Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0271] For example, when the computer 80 functions as the information processing device 1 according to the embodiment, the CPU 81 of the computer 80 executes programs loaded onto the RAM 82 to realize the functions of the processing unit 12. In addition, the HDD 84 stores data in the storage unit 11. The CPU 81 of the computer 80 reads and executes these programs from a recording medium 88, but as another example, the CPU 81 may obtain these programs from another device via the network N.

[0272] [7. Other] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0273] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0274] For example, the information processing device 1 described above may be realized by a terminal device and a server computer, or may be realized by multiple server computers. Furthermore, depending on the function, the configuration can be flexibly changed, such as by calling an external platform using an API or network computing.

[0275] Furthermore, the above-described embodiments and modifications can be combined as appropriate within the scope of not causing any contradiction in the processing content.

[0276] [8. Effects] As described above, the information processing device 1 according to the embodiment includes the object-related information acquisition unit 40, the user information acquisition unit 41, and the generation unit 34. The object-related information acquisition unit 40 acquires object information including information about a specific object and object impression information indicating an impression that a user U is estimated to have of the specific object. The user information acquisition unit 41 acquires user impression information indicating an impression that the user U is determined to have of the object information. The generation unit 34 generates improvement information including information indicating improvements to the object information, based on the object information and object impression information acquired by the object-related information acquisition unit 40 and the user impression information acquired by the user information acquisition unit 41. This enables the information processing device 1 to improve the convenience of a provider of a specific object.

[0277] The information processing device 1 also includes a receiving unit 30 that receives improvement directionality information indicating a direction for improving the target information, and the generating unit 34 generates the improvement information based on information that further includes the improvement directionality information received by the receiving unit 30. This enables the information processing device 1 to further improve the convenience for the provider of the identified target.

[0278] Furthermore, the directionality of improvement of the target information includes a directionality of improvement toward an impression that is estimated to be held by the user U and a directionality of improvement toward an impression that is determined to be held by the user U. This allows the information processing device 1 to further improve the convenience of the provider of the specified target.

[0279] Furthermore, the generation unit 34 generates the improvement information using the generation AI, which allows the information processing device 1 to generate the improvement information with high accuracy.

[0280] The information processing device 1 also includes a determination unit 33 that determines the impression that the user U has based on the information about the specific target of the user U. This allows the information processing device 1 to further improve the convenience of the provider of the specific target.

[0281] Furthermore, the determination unit 33 determines the impression that the user U has based on the message posted by the user U regarding the specific target. This allows the information processing device 1 to further improve the convenience of the provider of the specific target.

[0282] The information processing device 1 includes an estimation unit 32 that estimates, based on the target information, the impression that the user U is likely to have. This allows the information processing device 1 to further improve convenience for the provider of the identified target.

[0283] Furthermore, the target information is an advertisement for a specific target, which allows the information processing device 1 to further improve convenience for the provider of the specific target.

[0284] Furthermore, the target information is a catchphrase of the specific target. This allows the information processing device 1 to further improve convenience for the provider of the specific target.

[0285] The generation unit 34 also inputs information including the target information, the target impression information, the user impression information, and instruction information corresponding to the improvement direction information to the generation AI as input information, and causes the generation AI to output improvement information. This allows the information processing device 1 to further improve the convenience of the provider of the identified target.

[0286] The above describes the embodiments of the present application in detail based on the drawings, but this is merely an example, and the present invention can be implemented in other forms that include the embodiments described in the Disclosure of the Invention section and that have been modified and improved in various ways based on the knowledge of those skilled in the art.

[0287] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit." For example, an acquisition unit can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0288] 1,4 Information processing equipment 2,3 Terminal equipment 10. Communications Department 11 Storage section 12 Processing section 20 User information storage unit 30 Reception 31 Acquisition Department 32 Estimation part 33 Judgment section 34 Generation part 35 Providing Department 40 Subject-related information acquisition unit 41 User information acquisition unit 100 Information Processing Systems N Network

Claims

1. an object-related information acquisition unit that acquires object information including information on a specific object and object impression information indicating an impression that a user is estimated to have about the specific object; a user information acquisition unit that acquires user impression information indicating an impression determined to have been held by the user with respect to the target information; a generation unit that generates improvement information including information indicating an improvement content related to the target information, based on the target information and the target impression information acquired by the target-related information acquisition unit and the user impression information acquired by the user information acquisition unit.

1. An information processing device comprising:

2. a receiving unit that receives improvement direction information that indicates a direction of improvement of the target information; The generation unit generating the improvement information based on information further including the improvement directionality information received by the receiving unit; 2. The information processing apparatus according to claim 1, wherein:

3. The direction of improvement of the target information is as follows: The direction of improvement toward the impression that is estimated to be held by the user and the direction of improvement toward the impression that is determined to be held by the user are included.

3. The information processing apparatus according to claim 2, wherein:

4. The generation unit Generate the improvement information using the generation AI 4. The information processing device according to claim 1, wherein the information processing device is a computer.

5. a determination unit that determines an impression that the user has based on information about the specific target that the user has 4. The information processing device according to claim 1, wherein the information processing device is a computer.

6. The determination unit Determining an impression of the user based on a message posted by the user regarding the specific target 6. The information processing apparatus according to claim 5,

7. An estimation unit that estimates an impression that the user is likely to have based on the target information.

4. The information processing device according to claim 1, wherein the information processing device is a computer.

8. The target information is The targeted advertisement 4. The information processing device according to claim 1, wherein the information processing device is a computer.

9. The target information is A catchphrase for the specific target 9. The information processing apparatus according to claim 8,

10. The generation unit Inputting information including the target information, the target impression information, the user impression information, and instruction information corresponding to the improvement directionality information into a generation AI as input information, and causing the generation AI to output the improvement information.

3. The information processing apparatus according to claim 2, wherein:

11. 1. A computer-implemented information processing method, comprising: an object-related information acquisition step of acquiring object information including information on a specific object and object impression information indicating an impression that a user is estimated to have about the specific object; a user information acquiring step of acquiring user impression information indicating an impression determined to have been held by the user with respect to the target information; a generating step of generating improvement information including information indicating an improvement content related to the target information based on the target information and the target impression information acquired by the target-related information acquiring step and the user impression information acquired by the user information acquiring step.

1. An information processing method comprising:

12. an object-related information acquisition step of acquiring object information including information on a specific object and object impression information indicating an impression that a user is estimated to have about the specific object; a user information acquisition step of acquiring user impression information indicating an impression determined to have been held by the user with respect to the target information; a generating step of generating improvement information including information indicating an improvement content related to the target information, based on the target information and the target impression information acquired by the target-related information acquiring step and the user impression information acquired by the user information acquiring step. An information processing program characterized by:

Images