Personality evaluation system

Abstract

To provide a personality evaluation system S that can appropriately evaluate the personality of a user U.SOLUTION: A personality evaluation system S of the present invention comprises a service management device 1 that performs provision and management of a conversation service for a user terminal 3, and generative AI 2 that generates virtual simulated personality C corresponding to a user U who uses the user terminal 3. The generative AI 2 has a simulated personality generation unit 21 that generates simulated personality C reflecting the personality of the user U on the basis of the conversation information of the user U, a skill correction unit 22 that makes a correction on the simulated personality C in correspondence with the skill of the user U; and a sense-of-values correction unit 23 that makes a correction on the simulated personality C in correspondence with a threshold of the user U's sense of values obtained from the conversation information of the user U.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a personality evaluation system for evaluating a user's personality. [Background technology]

[0002] Conventionally, so-called chat services have been provided that allow real-time text communication using information processing terminals such as smartphones, tablets, or computers. Chat modes include user-to-user chat between users on a one-to-one basis, intra-community chat between multiple users belonging to a specific community, and inter-community chat. This type of chat service technology is disclosed, for example, in Patent Document 1.

[0003] In chat services, users typically go through a predetermined approval process, such as friend registration or community registration, before starting a chat, and the number of users is then increased. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 5175402 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the past, the number of users increased without the involvement of the service provider, and there was concern that an unspecified number of users would, whether consciously or unconsciously, engage in acts such as criticizing, slandering, or libeling others, or that users who were put off by this type of behavior would stop using the chat service, leading to a depopulation of the chat service itself.

[0006] One possible solution in this case would be for the service provider to constantly check and monitor the content posted in chat to prevent the above problems from occurring, but this would not only be time-consuming and costly, but would not be able to prevent all problems.

[0007] If service providers knew the personalities, such as the character and temperament, of all users of chat services, it would be possible to prevent trouble between users by, for example, warning users who are likely to cause trouble. It would also be possible to revitalize sparsely populated communities by encouraging users who are good at livening up the atmosphere to join. In other words, if service providers knew the personalities of all users of chat services, they could achieve both smooth operation and widespread adoption of chat services without having to constantly check and monitor the content posted in chat rooms. [Means for solving the problem]

[0008] The present invention has been made in view of the current situation as described above, and has as its technical object to provide a personality evaluation system that appropriately evaluates the personality of a user.

[0009] The personality evaluation system of the present invention comprises a service management device that provides and manages conversation services for a user terminal, and an imitation personality generation device that generates a virtual imitation personality corresponding to a user who uses the user terminal, and the imitation personality generation device has an imitation personality generation unit that generates an imitation personality that reflects the personality of the user based on the user's conversation information, a skill correction unit that applies corrections to the imitation personality in accordance with the skills possessed by the user, and a value correction unit that applies corrections to the imitation personality in accordance with a threshold value of the user's values ​​obtained from the user's conversation information.

[0010] In the personality evaluation system of the present invention, the imitation personality generation device may perform a free conversation style personality diagnosis for each of the user and the imitation personality, and the service management device may compare the personality diagnosis information of the user with the personality diagnosis information of the imitation personality and determine the imitation accuracy of the imitation personality from the degree of similarity.

[0011] In the personality evaluation system of the present invention, the service management device may have a number of imitative personalities for each user, and may have the imitative personalities evaluate the behavior of a given user obtained from the user's conversation information, and set the evaluation result as the characteristic waveform of the given user.

[0012] In the personality evaluation system of the present invention, the service management device may compare the characteristic waveform of the given user with the characteristic waveform of another given user and determine whether the two users are a match based on the degree of similarity.

[0013] In the personality evaluation system of the present invention, the service management device may set a collection of characteristic waveforms of multiple arbitrary users as a characteristic waveform of a specific community, compare the characteristic waveform of the arbitrary user with the characteristic waveform of the specific community, and determine whether the arbitrary user is matched to the specific community based on the similarity.

[0014] In the personality evaluation system of the present invention, the service management device may set a collection of characteristic waveforms of multiple arbitrary users as a characteristic waveform of a specific community, compare the characteristic waveform of the specific community with the characteristic waveform of another specific community, and determine whether the two specific communities match based on the similarity. [Effects of the Invention]

[0015] According to the present invention, the personality of each user using a conversation service can be easily understood from the corresponding imitation personality. Therefore, for example, by restricting the use of the conversation service for users who are likely to cause trouble, trouble between users can be prevented. Furthermore, by encouraging users who are good at livening up the atmosphere to participate in a community that is becoming increasingly depopulated, the community can be revitalized. In short, the smooth operation and widespread use of the conversation service can be achieved without the service provider having to constantly check and monitor the content of conversations. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram illustrating a configuration of a personality evaluation system according to an embodiment. [Figure 2] FIG. 2 is a conceptual diagram illustrating a hardware configuration of a service management device. [Figure 3] FIG. 2 is a conceptual diagram showing the hardware configuration of a user terminal. [Figure 4] FIG. 10 is a sequence diagram showing an example of generating an imitation personality. [Figure 5] FIG. 10 is a sequence diagram showing an example of improving the accuracy of an imitation personality. [Figure 6] 10 is a flowchart showing an example of improving the accuracy of a personality evaluation of a user. [Figure 7] FIG. 10 is a diagram showing a value waveform of a user evaluated by an imitation personality. [Figure 8] 10 is a flowchart illustrating an example of an approval process for determining whether or not to allow a user to participate in a specific community. DETAILED DESCRIPTION OF THE INVENTION

[0017] Next, embodiments embodying the present invention will be described with reference to the drawings. While the drawings show preferred embodiments, the present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0018] First, an overview of a personality evaluation system S (hereinafter simply referred to as "system S") according to an embodiment will be described with reference to Fig. 1 etc. System S evaluates the personality of user U using a generative system AI2 (generative artificial intelligence) as an imitation personality generation device based on the content of the user U's conversation in a chat service, which is an example of a conversation service.

[0019] As shown in FIG. 1, a system S includes a service management device 1 operated by a service provider of the system S, a generation system AI 2, and a plurality of user terminals 3.

[0020] The service management device 1 is configured with one or more computers and functions, for example, as a web server on the communication network 4. The service management device 1 of the embodiment is configured to provide and manage chat services to each user terminal 3. The service management device 1 is configured to be communicatively connected to each user terminal 3 via the communication network 4, and is capable of exchanging various information with each user terminal 3.

[0021] Generative AI2 utilizes methods including well-known AI programs, AI libraries, or AI platforms, such as machine learning and deep learning. Examples of AI technologies that can be used for text data, audio data, or video data include well-known models and methods such as hidden Markov models, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory (LSTMs).

[0022] The generative AI 2 of the embodiment uses a learning model based on a large-scale language model (LLM) such as ChatGPT. The generative AI 2 is configured to be communicatively connected to the service management device 1 via an API (Application Programming Interface).

[0023] Based on the user behavior information 7 of each user U, the generative AI 2 extracts emotions such as joy, anger, sadness, and happiness, positive actions such as gratitude and praise, and negative actions such as slander. The generative AI 2 can then use the extracted results as indicators to perform a personality evaluation of the user U. For example, a user U who uses the chat app 5 infrequently or who engages in negative actions such as slander against other users U is given a lower evaluation. A user U who livens up the atmosphere, engages in positive actions such as gratitude and praise, and uses language that is well-received by other users U is given a higher evaluation. The user behavior information 7 includes, for example, various texts, images, and files such as chats, and may include not only text but also audio, images, and video.

[0024] The generation system AI2 includes an imitation personality generation unit 21 that generates an imitation personality C that reflects the personality of user U based on user behavior information 7 of user U, a skill correction unit 22 that applies corrections to the imitation personality C in accordance with the skills possessed by user U, and a value correction unit 23 that applies corrections to the imitation personality C in accordance with a threshold value of user U's values ​​obtained from user behavior information 7 of user U.

[0025] The mimic personality generation unit 21 generates each mimic personality C (the program that constitutes it) by analyzing the user behavior information 7 of each user U collected and stored in the user behavior information DB 19 of the storage unit 13, which will be described later. Each mimic personality C is constructed by a computer program, behaves as if it were a virtual human being similar to a real human being, and has the personality and emotions that imitate the corresponding user U. The emotions of each mimic personality C are set to change in the same way as the emotions of the corresponding user U. In other words, each mimic personality C is set to simulate (or emulate) the emotions of the corresponding user U.

[0026] The mimic personality C is an artificial intelligence (AI) that mimics the personality, emotions, and emotional changes of the user U, and specifically includes a program that executes an algorithm and data processed by the program. The mimic personality C has algorithms corresponding to a learning function that learns about the user U based on the user behavior information 7, an emotion detection function that estimates (detects) the emotions of the user U, and a conversation execution function that conducts a conversation with the user U. Each mimic personality C is executed and processed by the mimic personality execution unit 17 (details of which will be described later) of the information processing unit 12 in the service management device 1.

[0027] The skill correction unit 22 executes a process of correcting the tendency of the imitation personality C for each user U based on the skill (e.g., occupation, etc.) of the user U. The skill of the user U means a high level of ability cultivated through training and learning. In the embodiment, the skill is inferred from the user U's occupation, but the skill may also be determined based on the user U's other abilities.

[0028] The value correction unit 23 analyzes the thresholds of each judgment (value) held by each user U, such as good and evil, likes and dislikes, gains and losses, interests, pain and pleasure, difficulty and difficulty, yes and no, pleasure and discomfort, and right and wrong, based on the user behavior information 7 of the user U, and performs processing to set the analysis results as the thresholds of each judgment in the corresponding imitated personality C.

[0029] Each user terminal 3 is an information processing terminal through which the user U who owns it receives chat services. Examples of the user terminal 3 include a smartphone, a tablet, or a computer. Each user terminal 3 is configured to be communicatively connected to the service management device 1 via a communication network 4, and is capable of exchanging various types of information with the service management device 1.

[0030] Each user terminal 3 can communicate with another terminal 3 via the service management device 1 and the communication network 4. When using the chat service for the first time, the user accesses the service management device 1 using the user terminal 3, downloads the chat application 5, and installs it on the user terminal 3.

[0031] Each user terminal 3 stores user information including personal data of the user U. Examples of user information include name, sex, date of birth (age), occupation, mobile phone number, chat service user ID 6 (see FIG. 2), community ID, etc. The communication network 4 is configured by, for example, the Internet, a mobile communication network, and base stations, but there are no restrictions on the type of communication protocol, or the type or scale of the network.

[0032] Fig. 2 is a conceptual diagram showing the hardware configuration of the service management device 1. As shown in Fig. 2, the service management device 1 includes a communication unit 11, an information processing unit 12, and a storage unit 13. The communication unit 11, the information processing unit 12, and the storage unit 13 are connected to each other via a system bus 14.

[0033] The communication unit 11 is, for example, an NIC for Ethernet or various communication modules such as a wired LAN or a wireless LAN, and has a function of connecting to the communication network 4 for communication.

[0034] The information processing unit 12 includes a RAM, a ROM, a CPU (not shown), etc. In accordance with a chat management program or the like stored in the storage unit 13, the information processing unit 12 executes, for example, a process of acquiring chat information, which is conversation information, from each user terminal 3 and recording it in the storage unit 13, and a process of transmitting it to other user terminals 3 belonging to a community or the like.

[0035] The information processing unit 12 includes a personality similarity calculation unit 15, a waveform similarity calculation unit 16, an imitation personality execution unit 17, and a value waveform generation unit 24. The personality similarity calculation unit 15 executes a process to compare the personality diagnostic information of user U and imitation personality C obtained from a conversation with the generative AI 2 to determine the degree of personality similarity, indicating the degree of match. As will be described in detail later, the value waveform generation unit 24 executes a process to calculate value waveforms G+(U or SC), G-(U or SC) as characteristic waveforms indicating the cumulative results of virtues and landmines for each imitation personality C for each user U or specific community SC. Furthermore, the waveform similarity calculation unit 16 executes a process to compare the value waveforms G+(U), G-(U) of each user U with the value waveforms G+(SC), G-(SC) of the specific community SC to determine the degree of match. The imitation personality execution unit 17 is responsible for executing the imitation personality C, which is a program including an artificial intelligence algorithm.

[0036] Here, the personality similarity is an index showing the degree of similarity between the personality diagnostic information of the user U and the imitation personality C. As the personality similarity, for example, the cosine similarity between the feature vector of the personality diagnostic information of the imitation personality C and the feature vector of the personality diagnostic information of the user U may be used. The cosine similarity evaluates the degree of similarity between pieces of information based on the closeness (angle) of the directions of the vectors of the pieces of information.

[0037] To explain in more detail, for example, the personality similarity calculation unit 15 first calculates the number a of all elements of the feature vector where both pieces of personality diagnostic information have a value of 1. Then, the personality similarity calculation unit 15 calculates the number b of all elements of the feature vector where either piece of personality diagnostic information has a value of 1. Then, the personality similarity calculation unit 15 divides the calculated number a by the number b to calculate the cosine similarity (a / b) between the feature vector of the personality diagnostic information of the imitated personality C and the feature vector of the personality diagnostic information of the user U. In this case, the maximum value of the personality similarity is "1" and the minimum value is "0".

[0038] Furthermore, waveform similarity is an index showing the degree of similarity between the value waveforms G+(U) and G-(U) between users U, the degree of similarity between the value waveforms G+(U) and G-(U) of user U and the value waveforms G+(SC) and G-(SC) of a specific community SC, or the degree of similarity between the value waveforms G+(SC) and G-(SC) of a specific community SC and the value waveforms G+(SC) and G-(SC) of another specific community SC. The waveform similarity may also be the cosine similarity between the feature vectors of the value waveforms.

[0039] The method for calculating the personality similarity and waveform similarity is not limited to cosine similarity, and other calculation functions or other methods such as template matching or pattern matching may be used. In other words, any method may be applied to calculate the personality similarity and waveform similarity.

[0040] The storage unit 13 is a non-volatile memory such as a hard disk drive (HDD), flash memory (solid state drive (SSD)), or other solid-state memory, and stores an operating system, a chat management program, etc. The storage unit 13 stores and manages the user ID 6 of a user U who uses the chat service, the community ID of the chat community, etc., and also stores user behavior information 7 as conversation information acquired from each user terminal 3 in association with the corresponding user ID 6.

[0041] As databases containing this type of data, the storage unit 13 of the embodiment has constructed a user information database 18 (hereinafter referred to as "user information DB18"), a user speech and behavior information database 19 (hereinafter referred to as "user speech and behavior information DB19"), and an imitation personality database 20 (hereinafter referred to as "imitation personality DB20"), etc. Note that each of the DBs 18 to 20 may be provided in a network server (not shown) on the communication network 4 separate from the service management device 1, or may be provided in a local server (not shown) separately connected to the service management device 1.

[0042] The various types of information stored and managed in the memory unit 13 are preferably kept (stored) on a blockchain 8 (see FIG. 1). At least the various types of information stored in the user information DB 18 of the memory unit 13 are preferably kept on the blockchain 8. The blockchain 8 (distributed ledger) is based on a P2P network consisting of multiple computer-related devices (nodes) connected to the communication network 4, and is constructed using distributed applications and smart contracts provided by, for example, Ethereum or Polygon.

[0043] When various types of information are stored on the blockchain 8, it is preferable that each user terminal 3 generates a transaction to the blockchain 8, and when the transaction is approved on the blockchain 8, the various types of information stored on the blockchain 8 are updated. The service management device 1 may function as a node that constitutes the blockchain 8. Note that as the blockchain 8, a public blockchain such as Ethereum or Polygon may be applied, or other private blockchains may be applied.

[0044] The user information DB 18 is a database that registers various information about a user U who has been issued a user account for the system S. In the user information DB 18, some of the user information described above, i.e., a user ID 6 for user identification, a password, the occupation of the user U, a community ID indicating the chat community to which the user U belongs, and the like, are registered in association with each user terminal 3.

[0045] The user behavior information DB19 is a database that collects and stores a large amount of user behavior information 7, such as various texts, images, and files of messages (conversations) when using the chat service, acquired from each user terminal 3, in association with each user ID 6. The user behavior information DB19 includes a table that uses the user ID 6 that identifies each individual user U as a primary key and compiles the user behavior information 7 acquired from each user terminal 3 as a history by appropriate time period (for example, every 24 hours (1 day), every hour, etc.). The imitation personality DB20 is a database that stores and associates multiple imitation personalities C generated by the generation system AI2 with each user ID 6.

[0046] In an embodiment, each imitation personality C (the program that constitutes it) is generated by the imitation personality generation unit 21 (see Figure 1) of the generation system AI2 analyzing the user behavior information 7 of each user U collected and stored in the user behavior information DB19 of the memory unit 13.

[0047] In this case, the imitation personality generation unit 21 of the generative AI 2 refers to, for example, the Big Five theory of psychology to extract emotions such as joy, anger, sadness, and happiness, positive actions such as gratitude and praise, and negative actions such as slander, based on the user behavior information 7 of each user U, and analyzes the character and emotions (so-called personality) of the user U. The Big Five theory classifies the personality of the user U into five characteristics. The five characteristics evaluated by the generative AI 2 (imitation personality generation unit 21) include agreeableness, conscientiousness, extroversion, neuroticism, and openness.

[0048] For example, agreeableness refers to the trait of striving for social harmony and individual adjustment. A user U with a high score on agreeableness is generally considered to be thoughtful, kind, generous, and trustworthy, and to be willing to compromise between the interests of others and his or her own. Conscientiousness refers to self-control, and a user U with a high score on conscientiousness is considered to have the trait of acting honestly and pursuing achievement against external expectations. Extroversion is the trait of gaining energy from diverse activities or external environments. A user U with a high score on extroversion is considered to like interacting with other users U, and to be passionate and active.

[0049] Neuroticism refers to a tendency to easily feel negative emotions such as anger, anxiety, or depression. A user U with a high score on neuroticism is understood to be emotionally reactive, vulnerable to stress, and susceptible to change depending on how they express their emotions. Openness is an evaluation of art, emotion, adventure, imagination, curiosity, and diverse experiences. A user U with a high score on openness is understood to be intellectually curious, sensitive to beauty, and willing to try new things.

[0050] The imitation personality generation unit 21 of the generative system AI2 assigns tendencies ranging from 0% to 100% to the five characteristics, indexes (e.g., scores) each characteristic of the user U based on such tendencies, and converts them into the corresponding characteristics of the imitation personality C.

[0051] Then, the skill correction unit 22 of the generation system AI2 corrects the tendency of each characteristic based on the skills (occupation in this embodiment) of each user U, reflects the corrected tendency of each characteristic as an index (e.g., a score, etc.) of each characteristic of the user U, and converts it into the corrected characteristics of the corresponding imitation personality C.

[0052] Users U (imitated personalities C) with similar distributions of indices (e.g., scores) calculated by the skill correction unit 22 of the generative AI 2 are evaluated as having similar (close) personalities. Users U (imitated personalities C) with completely different distributions of indices (e.g., scores) are evaluated as having different (distant) personalities.

[0053] The method for analyzing personality, etc. is not limited to the Big Five theory, but may also be the MBTI (Myers-Briggs Type Indicator), StrengthsFinder (registered trademark), or other analytical methods, which will be described later.

[0054] Furthermore, the value correction unit 23 of the generative AI2 analyzes the thresholds of each judgment (value) such as good and evil, likes and dislikes, gains and losses, interests, pain and pleasure, difficulty and difficulty, yes and no, pleasure and discomfort, and right and wrong based on the user behavior information 7 of each user U, and converts them into indices (for example, scores) and converts them into thresholds of each judgment in the corresponding imitative personality C.

[0055] In this case, too, users U (imitated personalities C) with similar distributions of indices (e.g., scores, etc.) calculated by the value correction unit 23 of the generative AI 2 are evaluated as having similar (close) senses of judgment (values). Users U (imitated personalities C) with completely different distributions of indices (e.g., scores, etc.) are evaluated as having different (distant) senses of judgment (values).

[0056] Each mimic personality C has a different threshold of values ​​(which can also be considered a judgment scale or standard). In other words, the thoughts and standards of each mimic personality C are individually different, just like those of an ordinary person. This was adopted in light of the lack of reliability of methods for quantifying human thoughts and standards of judgment using a single standard.

[0057] For example, depending on the level of tolerance for "not being punctual" or "speaking informally," one mimic personality C may give a zero rating (not see it as a problem), while another may give a negative rating or multiple negative ratings. In other words, it is up to each mimic personality C to decide whether to give a positive rating, negative rating, or zero rating to each user U's behavioral item. Roughly speaking, a positive rating is when the mimic personality C makes a favorable judgment, such as a positive act such as gratitude or praise, and a negative rating is when the mimic personality C makes an unfavorable judgment, such as a negative act such as slander.

[0058] The behavioral items for evaluating each user U are evoked by the generative AI 2 or any imitative personality C based on the user behavior information 7 of each user U, and the evoked behavioral items are set to be evaluated and judged as common items by all imitative personalities C. In the embodiment, a positive evaluation may be expressed as "virtue" and an accumulation of positive evaluations as "accumulating virtue," while a negative evaluation may be expressed as "landmine" and an accumulation of negative evaluations as "continuing to step on landmines."

[0059] In the system S of the embodiment, each mimicked personality C (the program constituting it) generated by the generation system AI2 is stored in the mimicked personality DB20 of the recording unit 13 in the service management device 1. Then, as will be described in detail later, the service management device 1 (mimetic personality execution unit 17) executes the program constituting each mimicked personality C. That is, each mimicked personality C operates by the service management device 1 (mimetic personality execution unit 17) executing the program constituting each mimicked personality C.

[0060] In this case, the service management device 1 detects and collects user behavior information 7 of each user U at a predetermined timing. The user behavior information 7 of each added or updated user U is also stored in the user behavior information DB 19 in association with each user ID 6. The generation system AI 2 (imitated personality generation unit 21) analyzes the user behavior information 7 of each added or updated user U and makes the corresponding imitated personality C learn from it, and the imitated personality execution unit 17 of the service management device 1 analyzes the user behavior information 7 of each added or updated user U and makes the corresponding imitated personality C learn from it.

[0061] Through such learning, each mimic personality C can more accurately imitate the personality, emotions, and judgment criteria (thresholds) of the corresponding user U, and can acquire an algorithm that reproduces the personality, emotions, and judgment criteria (thresholds) of the user U with high accuracy, specialized for that user U. The algorithm of each mimic personality C is improved and modified each time through repeated learning. As the learning of each mimic personality C progresses, the service management device 1 updates the mimic personality C and stores it in the mimic personality DB 20.

[0062] Fig. 3 is a conceptual diagram showing the hardware configuration of the user terminal 3. As shown in Fig. 3, the user terminal 3 has a CPU 31, memory 32, storage 33, communication IF 34, input device 35, display unit 36, input / output IF 37, etc. These are connected to each other via a system bus 38. The CPU 31 performs various calculations in accordance with programs and performs processing to realize the functions of the user terminal 3. The number of CPUs 31 may be one or more. When multiple CPUs 31 are used, these CPUs 31 may execute processing simultaneously or sequentially.

[0063] The memory 32 functions as a work area when the CPU 31 executes processing. The memory 32 includes, for example, a RAM and a ROM. The storage 33 stores various programs and data and includes, for example, an SSD and / or an HDD. The storage 33 stores programs for realizing the functions of the user terminal 3, such as the chat application 5 provided by the service management device 1.

[0064] The communication IF 34 communicates with other devices in accordance with a predetermined communication standard (e.g., Ethernet). The communication IF 34 includes, for example, a NIC. The input device 35 inputs information in response to an operation by the user U. The input device 35 includes, for example, at least one of a touch sensor, a keyboard, a keypad, a mouse, and a microphone.

[0065] The display unit 36 ​​displays various types of information. The display unit 36 ​​includes, for example, an LCD. The display unit 36 ​​may be integrated with a touch sensor to form a touch screen. The input / output IF 37 is an interface for connecting the input device 35, a camera, etc.

[0066] Next, an example of the operation of the personality evaluation system S according to the embodiment, in which an imitation personality Cn (subscript n is an integer of 1 or more) corresponding to an arbitrary user Un is described with reference to FIG.

[0067] As shown in Figure 4, when a user Un launches a chat application 5 on his / her own user terminal 3 and starts chatting or other behavior (behavior from which information can be obtained) with another user U (step S01: YES), the service management device 1 collects and stores user behavior information 7n such as messages from the user terminal 3n of user Un in the user behavior information DB19, associating it with the user ID 6n (step S02).

[0068] Next, when a predetermined amount of user behavior information 7n is collected (step S03: YES), the service management device 1 sends a request to generate an imitation personality Cn including the user ID 6n and the user behavior information 7n to the generation system AI2 via the API (step S04).

[0069] Upon receiving a request to generate an imitation personality Cn from the service management device 1, the generation system AI2 causes the imitation personality generation unit 21 to index (e.g., score) each characteristic of the user Un through analysis based on the user behavior information 7n, and generates an imitation personality Cn (program) in response to the request (step S05). Next, the skill correction unit 22 of the generation system AI2 reflects corrections based on the user Un's occupation in the imitation personality Cn (step S06). Then, the generation system AI2 causes the value correction unit 23 to analyze each judgment threshold of the user Un based on the user behavior information 7n, and index (e.g., score), and reflects the index in the imitation personality Cn (step S07).

[0070] When the generation of the mimic personality Cn by the generation system AI2 is completed, a notification of the completion of the generation of the mimic personality Cn, including the mimic personality Cn (program), is sent from the generation system AI2 to the service management device 1 via the API (step S08). Next, the service management device 1, which has received the notification of the completion of the generation of the mimic personality Cn from the generation system AI2, associates the mimic personality Cn with the user ID 6n, and registers and stores it in the mimic personality DB20 of the storage unit 13 (step S09).

[0071] According to the above control, the system includes a service management device 1 that provides and manages conversation services for user terminals 3, and an imitation personality generation device 2 (generative AI) that generates a virtual imitation personality C corresponding to a user U using the user terminal 3. The imitation personality generation device 2 includes an imitation personality generation unit 21 that generates an imitation personality C that reflects the personality of user U based on the user U's conversation information, a skill correction unit 22 that corrects the imitation personality C in accordance with the user U's skills, and a value correction unit 23 that corrects the imitation personality C in accordance with the user U's value threshold obtained from the user U's conversation information. Therefore, the personality of each user U using the conversation service can be easily understood from the corresponding imitation personality C. Therefore, for example, by restricting the use of the conversation service for users U who are likely to cause trouble, trouble between users U can be prevented. Furthermore, by encouraging users U who are good at livening up the atmosphere to participate in a community that is becoming increasingly depopulated, the community can be revitalized. In short, the smooth operation and widespread use of conversation services can be achieved without the need for service providers to constantly check and monitor the content of conversations.

[0072] Next, an example of the operation of the personality evaluation system S according to the embodiment, in which the imitation personality Cn corresponding to an arbitrary user Un is improved in accuracy, will be described with reference to Fig. 5. In this case, it is assumed that the imitation personality Cn corresponding to the user Un has already been stored in the storage unit 13 (imitation personality DB 20) of the service management device 1.

[0073] 5, while a user Un is engaged in an action such as chatting with another user U (an action from which information can be obtained) (step T01: YES), the service management device 1 determines whether a personality diagnosis of the user Un has been executed to a high degree of accuracy until a result that imitates the personality diagnosis is obtained (step T02). If a personality diagnosis of the user Un has not been executed (step T02: NO), the service management device 1 transmits a request to execute a free-conversation style personality diagnosis for the user Un to the generation system AI2 via the API (step T03).

[0074] Upon receiving the request to execute a personality diagnosis from the service management device 1, the generation system AI2 generates a number of questions in a free conversation format in response to the request to execute a personality diagnosis, and carries out a conversation including the questions with the user Un (step T04).

[0075] Personality diagnosis by generative AI2 is not conducted in the conventional format of "user Un choosing from several options" (static personality diagnosis), but in a format that encourages smooth judgment by user Un through chat (conversation) between user Un and generative AI2. This is because in static personality diagnosis, questions are set in advance so that "user Un chooses from several options," which leaves room for consideration in terms of inconsistency and lack of accuracy in user Un's answers.

[0076] In response to this, for example, generative AI2 continues the chat without interruption and generates questions in line with the content and flow of the chat while adapting to user Un's responses. In other words, generative AI2 behaves dynamically in line with user Un's responses and performs a personality diagnosis (dynamic personality diagnosis) for user Un. If generative AI2 performs such a dynamic personality diagnosis, it can obtain all the information necessary for a personality diagnosis for user Un from the natural flow of conversation, and the obtained information will accurately reflect user Un's personality.

[0077] The personality test used in the embodiment is, for example, the Myers-Briggs Type Indicator (MBTI). The MBTI diagnostic method is already known, so a detailed explanation will be omitted. The personality test used here is not limited to the MBTI, and may be the aforementioned Big Five theory, StrengthsFinder (registered trademark), or other personality test methods.

[0078] When the personality diagnosis of user Un by generation system AI2 is completed, a personality diagnosis completion notice including the personality diagnosis information of user Un is sent from generation system AI2 to service management device 1 via API (step T05). Next, upon receiving the personality diagnosis completion notice from generation system AI2, service management device 1 links the personality diagnosis information of user Un to user ID 6n and registers and stores it in memory unit 13 (step T06).

[0079] Next, the service management device 1 activates the imitation personality Cn corresponding to the user Un and sends a request to the generation system AI2 via the API to execute a personality diagnosis in a free conversation format for the imitation personality Cn (step T07). In this case, the execution process for the imitation personality Cn is performed by the imitation personality execution unit 17. Upon receiving the personality diagnosis execution request from the service management device 1, the generation system AI2 generates a number of questions in a free conversation format in response to the personality diagnosis execution request and executes a conversation including the questions with the imitation personality Cn (step T08). In step T08, the generation system AI2 executes a dynamic personality diagnosis for the imitation personality Cn, as in step T04.

[0080] When the personality diagnosis of the imitated personality Cn by the generative system AI2 is completed, a personality diagnosis completion notice including the personality diagnosis information of the imitated personality Cn is sent from the generative system AI2 to the service management device 1 via the API (step T09). Next, the service management device 1 (personality similarity calculation unit 15) that has received the personality diagnosis completion notice from the generative system AI2 links the personality diagnosis information of the imitated personality Cn to the user ID 6n, and registers and stores it in the memory unit 13 (step T10).

[0081] Next, the service management device 1 (personality similarity calculation unit 15) reads out the personality diagnosis information of user Un and the personality diagnosis information of the imitated personality Cn from the storage unit 13, compares them, and calculates the personality similarity of the imitated personality C to user U (step T11). If the personality similarity of the imitated personality C to user U exceeds a predetermined threshold (step T11: YES), it can be said that the imitated personality Cn imitates the personality, emotions, and judgment criteria (thresholds) of user Un with high accuracy and has acquired an algorithm that reproduces the personality, emotions, and judgment criteria (thresholds) with high accuracy, specialized for user Un. Therefore, it is considered that the imitated personality Cn is complete, and the precision improvement process for the imitated personality Cn is terminated.

[0082] If the personality similarity of the imitated personality C to the user U falls below a predetermined threshold (step T11: NO), the service management device 1 sends an update request for the imitated personality Cn, including the added or updated user behavior information 7n and user ID 6n of the user Un, to the generation system AI2 via the API (step T12).

[0083] Upon receiving the request to update the mimicked personality Cn from the service management device 1, the generation system AI2 (the mimicked personality generation unit 21) analyzes the user behavior information 7n to index (e.g., score) each characteristic of the user Un in response to the request to update the mimicked personality Cn, and updates the mimicked personality Cn (the program that constitutes it) (step T13). Then, the skill correction unit 22 of the generation system AI2 reflects corrections based on the user Un's skills (e.g., occupation) in the mimicked personality Cn (step T14). Furthermore, the value correction unit 23 of the generation system AI2 analyzes the thresholds of each judgment of the user Un based on the added or updated user behavior information 7n, indexes (e.g., score), and reflects them in the mimicked personality Cn (step T15).

[0084] When the generation system AI2 completes updating the mimicked personality Cn, the generation system AI2 transmits a notification of completion of updating the mimicked personality Cn, including the mimicked personality Cn (program), to the service management device 1 via the API (step T16). Next, the service management device 1, which has received the notification of completion of updating the mimicked personality Cn from the generation system AI2, associates the mimicked personality Cn with the user ID 6n, and registers and stores the same in the mimicked personality DB 20 of the storage unit 13 (step T17).

[0085] Thereafter, the service management device 1 returns to step T07 and sends a request to the generation system AI2 via the API to perform a personality diagnosis in a free conversation format for the updated imitation personality Cn, and thereafter, the processing of steps T08 to T16 is executed in a loop until the personality similarity of the updated imitation personality Cn to the user Un exceeds a predetermined threshold.

[0086] According to the above control, the imitation personality generation device 2 (generation system AI) performs a free conversation style personality diagnosis for each of the user U and the imitation personality C, and the service management device 1 compares the personality diagnosis information of the user U with the personality diagnosis information of the imitation personality C and determines the imitation accuracy of the imitation personality C from the personality similarity, thereby achieving the effect of generating an imitation personality C that imitates the personality of the user U with higher accuracy, and being able to reproduce the personality of the user U with high accuracy, specialized for the user U.

[0087] Next, an example of the operation of the personality evaluation system S according to the embodiment, in which the personality evaluation of an arbitrary user Un is made more accurate, will be described with reference to FIGS.

[0088] 6, while user Un is chatting with another user U (step E01: YES), the service management device 1 (imitated personality execution unit 17) activates all imitated personalities C1 to C(n-1) stored in the imitated personality DB 20 and causes all imitated personalities C1 to C(n-1) to monitor user behavior information 7n of user Un (step E02). Then, the service management device 1 (imitated personality execution unit 17) causes all imitated personalities C1 to C(n-1) to evaluate a large number of behavior items recalled from the user behavior information 7n as "virtuous," "a landmine," or "no problem," and accumulates these evaluations for each imitated personality C1 to C(n-1) (step E03).

[0089] As mentioned above, the thinking and judgment criteria of each of the imitated personalities C1 to C(n-1) are individually different, just like humans. Therefore, whether to give a positive evaluation, a negative evaluation, or a zero evaluation to each of the user U's behavioral items depends on each imitated personality C (each of the imitated personalities C1 to C(n-1) makes a different judgment).

[0090] When evaluations by the imitated personalities C1 to C(n-1) for all recalled speech and behavior items are completed (step E04: YES), the value waveform generation unit 24 of the service management device 1 calculates a value waveform (which can also be called a value graph) as a characteristic waveform showing the cumulative results of virtues and landmines for each of the imitated personalities C1 to C(n-1) for user Un (step E05). In this embodiment, the value waveform is represented by a combination of a virtue waveform G+(Un) and a landmine waveform G-(Un). Figure 7 shows an example of the value waveforms G+(Un) and G-(Un) for user Un evaluated by the imitated personalities C1 to C(n-1). The positive side of the vertical axis of the value waveforms G+(Un) and G-(Un) represents the cumulative value of virtues, and the negative side represents the cumulative value of landmines.

[0091] By creating the value waveforms G+(U) and G-(U) in this manner, it is possible to project onto the value waveforms G+(U) and G-(U) how user U is evaluated by each of the imitation personalities C1 to Cn-1, thereby enabling a more accurate understanding of the surface and inner personality of each user U. The surface and inner personality of each user U can be visualized with high reliability. Users U whose value waveforms G+(U) and G-(U) have similar shapes can be evaluated as having similar (close) values. Users U whose value waveforms G+(U) and G-(U) have similar shapes can be expressed as having a high synchronization rate (which can also be said to be a similarity rate). Users U whose value waveforms G+(U) and G-(U) have completely different shapes can be evaluated as having different values. Users U whose value waveforms G+(U) and G-(U) have completely different shapes can be expressed as having a low synchronization rate.

[0092] Note that the value waveforms G+(U) and G-(U) described above are not the only ones that show the cumulative results of the virtues and landmines determined for user U for each imitation personality C. Instead of the value waveforms, a value color chart may be employed that uses colors such as "red" associated with victory, passion, love, etc., "yellow" associated with cheerfulness, youth, joy, etc., "green" associated with nature, harmony, peace, etc., and "blue" associated with quietness, intelligence, coldness, etc., to express the cumulative results of the virtues and landmines determined for user U for each imitation personality C in a gradational manner.

[0093] Next, an example of an approval process for determining whether or not to allow an arbitrary user Un to participate in a specific community SC to which users U1 to U(n-1) belong, among the operations of the personality evaluation system S according to the embodiment, will be described with reference to Fig. 8. Note that the specific community SC may be any of various communities, such as a corporate community, a fan community, an online salon, a knowledge community, a hobby community, or a local community.

[0094] Here, the graph shown in Figure 7 is obtained in advance as the value waveforms G+(Un) and G-(Un) of user Un. Furthermore, to represent the values ​​held by a collection of users U1 to U(n-1) in a specific community SC, the value waveforms G+(SC) and G-(SC) of the specific community SC are used, which are obtained by combining the value waveforms G+(U) and G-(U) of each user U. The value waveforms G+(SC) and G-(SC) correspond to the characteristic waveforms of the specific community SC.

[0095] The value waveforms G+(SC), G-(SC) of a specific community SC are calculated by the service management device 1 (value waveform generation unit 24) based on the value waveforms G+(U), G-(U) of each user U and stored in the storage unit 13. As an example of calculating the value waveforms G+(SC), G-(SC) of a specific community SC, for example, the value waveform G+(SC), G-(SC) can be calculated by taking the mode, median, or average of the cumulative values ​​for each imitation personality C in the virtue waveform G+(U) of each user U to calculate the virtue waveform G+(SC), and the mode, median, or average of the cumulative values ​​for each imitation personality C in the mine waveform G-(U) of each user U to calculate the mine waveform G-(SC), thereby obtaining the value waveforms G+(SC), G-(SC) of the specific community SC.

[0096] 8, when user Un starts chat application 5 on his / her user terminal 3 and applies to join specific community SC (step P01: YES), service management device 1 (waveform similarity calculation unit 16) reads out value waveforms G+(Un), G-(Un) of user Un and value waveforms G+(SC), G-(SC) of specific community SC from storage unit 13 (step P02). Next, service management device 1 (waveform similarity calculation unit 16) compares value waveforms G+(Un), G-(Un) of user Un with value waveforms G+(SC), G-(SC) of specific community SC to calculate waveform similarity of user Un to specific community SC (step P03).

[0097] Next, if the waveform similarity of user Un to specific community SC exceeds a predetermined threshold (step P04: YES), user Un's value waveforms G+(Un), G-(Un) are evaluated as being similar (high synchronization rate) to the value waveforms G+(SC), G-(SC) of specific community SC, and as being compatible (matching). Then, the service management device 1 (waveform similarity calculation unit 16) issues a command to user terminal 3 (chat application 5) of user Un to permit participation in specific community SC (step P05), enabling user Un to chat with other users U.

[0098] If the waveform similarity of user Un to the specific community SC falls below a predetermined threshold (step P04: NO), the value waveforms G+(Un), G-(Un) of user Un differ from the value waveforms G+(SC), G-(SC) of the specific community SC (the synchronization rate is low), and the values ​​are evaluated as incompatible (not matching). Therefore, the service management device 1 (waveform similarity calculation unit 16) issues a command to the user terminal 3 (chat application 5) of user Un to deny participation in the specific community SC (step P06), rendering user Un unable to participate in the specific community SC. While step P06 illustrates an example in which user Un's participation in the specific community is denied, the present invention is not limited to this. The service provider may set measures (such as imposing various restrictions on the specific community SC) based on the evaluation that values ​​are incompatible (not matching).

[0099] According to the above control, even if a user U who has different values ​​and is likely to cause trouble is invited, the user U can smoothly and reasonably refuse to participate in the specific community SC. This ensures that trouble between users U is prevented. Furthermore, since excellent users U who share similar values ​​and are considered to have good personalities gather together, the specific community SC can be easily activated and can develop into a safe and secure specific community SC. This also improves the reliability of the specific community SC.

[0100] Note that a user Un who has been denied participation in a specific community SC is not permanently barred from participating in the specific community SC. The value waveforms G+(Un) and G-(Un) of the user Un will naturally change depending on the accumulation of speech and behavior after the denial of participation. As a result, if the value waveforms G+(Un) and G-(Un) of the user Un become similar to the value waveforms G+(SC) and G-(SC) of the specific community SC, and the values ​​are evaluated as being compatible (similar, close), the user will be able to participate in the specific community SC.

[0101] In the embodiment shown in Figure 8, an example of a comparison (approval process) of the degree of similarity in the relationship between user U and a specific community SC is shown, but this is not limited to this, and a comparison of the degree of similarity in the relationship between user U and another user U can also be realized in the same way as the embodiment shown in Figure 8, and a comparison of the degree of similarity in the relationship between a specific community SC and another specific community SC can also be realized in the same way as the embodiment shown in Figure 8.

[0102] For example, when comparing the degree of similarity between a specific community SC and another specific community SC, if the values ​​of the two communities match (the degree of similarity is high), it is possible to recommend the other's specific community SC when searching, for example, and form a mutual partnership. Conversely, if the values ​​of the two communities do not match (the degree of similarity is low), it is possible to prevent the other's specific community SC from being introduced even when searching, for example.

[0103] The configuration of each part in the present invention is not limited to the illustrated embodiment, and various modifications are possible within the scope of the present invention. [Explanation of symbols]

[0104] C,C(n-1),Cn imitative personality G+(U),G+(Un) Virtue waveform G-(U),G-(Un) Mine waveform S Personality Assessment System SC Specific Community U,U(n-1),Un users 1 Service Management Device 2 Generative AI 3,3n user terminal 4. Communication Network 5. Chat Apps 6,6n User ID 7,7n User behavior information 11 Communications Department 12 Information Processing Department 13 Storage section 14,38 System Bus 15 Personality similarity calculation unit 16 Waveform similarity calculation unit 17 Imitation Personality Executive Department 18 User Information DB 19 User behavior information DB 20 Imitation Personality DB 21 Imitation Personality Generation Department 22 Skill Correction Section 23 Values ​​Correction Department 24 Value Waveform Generator 31 CPU 32 memory 33 Storage 34 Communication Interface 35 Input Devices 36 Display section 37 Input / Output Interface

Claims

1. The system includes a service management device that provides and manages conversation services for a user terminal, and an imitation personality generation device that generates a virtual imitation personality corresponding to a user who uses the user terminal, The imitation personality generation device has an imitation personality generation unit that generates an imitation personality that reflects the personality of the user based on conversation information of the user, a skill correction unit that corrects the imitation personality in accordance with the skills of the user, and a value correction unit that corrects the imitation personality in accordance with a threshold value of the user's value obtained from the conversation information of the user. Personality assessment system.

2. the imitation personality generation device executes a personality diagnosis in a free conversation format for each of the user and the imitation personality; the service management device compares the personality diagnostic information of the user with the personality diagnostic information of the imitated personality, and determines the imitation accuracy of the imitated personality based on the degree of similarity between the personality diagnostic information of the user and the imitated personality; 2. A personality evaluation system according to claim 1.

3. the service management device has a number of the mimic personalities for each user; having the plurality of imitative personalities evaluate the speech and behavior of a given user obtained from conversation information of the given user, and setting the evaluation result as a characteristic waveform of the given user; 3. A personality evaluation system according to claim 2.

4. the service management device compares the characteristic waveform of the given user with the characteristic waveform of another given user and determines whether the two users are a match based on the degree of similarity; 4. A personality evaluation system according to claim 3.

5. the service management device sets a collection of characteristic waveforms of a plurality of arbitrary users as a characteristic waveform of a specific community; comparing the characteristic waveform of the given user with the characteristic waveform of the specific community, and determining whether the given user matches the specific community based on the degree of similarity; 4. A personality evaluation system according to claim 3.

6. the service management device sets a collection of characteristic waveforms of a plurality of arbitrary users as a characteristic waveform of a specific community; comparing the characteristic waveform of the specific community with the characteristic waveform of another specific community, and determining whether the two specific communities match based on the degree of similarity; 4. A personality evaluation system according to claim 3.

Images