Information processing device, information processing method, and program

Abstract

To provide feedback information on the behaviors performed by the person being evaluated, taking into account their behavioral characteristics. [Solution] The information processing device comprises: a first acquisition unit that acquires evaluator information relating to evaluators who evaluate the actions of the person being evaluated; a second acquisition unit that acquires behavioral information relating to the actions of the person being evaluated; a first analysis unit that analyzes methods for changing the evaluator's behavior based on the evaluator information; a second analysis unit that analyzes methods for changing the person being evaluated based on the behavioral information; a feedback generation unit that generates feedback information for the person being evaluated based on the analysis results from the first and second analysis units; and an output unit that outputs the generated feedback information.

Description

Technical Field

[0001] The disclosed technology relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] Conventionally, when the content of presentation materials and the presenter's presentation content are different, a presentation coaching system that notifies the presenter of the different parts is known (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in recent years, efforts to utilize a behavioral science approach based on the theory of behavioral science that scientifically studies human behavior in service development have spread in various fields such as public policy, medicine, retail, and education. By using the behavioral science approach, it is possible to provide services optimized for users. That is, by using the behavioral science approach, users can receive the provision of various information in a manner that is, for example, simpler, more efficient, or less likely to be overlooked according to their respective behavioral characteristics.

[0005] However, in the presentation coaching system described in Patent Document 1, the presenter's behavioral characteristics are not considered, and the convenience is limited. Therefore, for example, it was not possible to output feedback information or the like for improving the presentation by considering the behavioral characteristics of the presenter, who is the evaluated person, in the presentation as an action.

[0006] Therefore, one of the objectives of the present invention is to provide feedback information on the actions taken by the person being evaluated, taking into account the behavioral characteristics of the person being evaluated. [Means for solving the problem]

[0007] An information processing device according to one aspect of the present invention comprises: a first acquisition unit that acquires evaluator information relating to an evaluator who evaluates the actions of a person being evaluated; a second acquisition unit that acquires behavioral information relating to the actions of a person being evaluated; a first analysis unit that analyzes a method for changing the evaluator's behavior based on the evaluator information; a second analysis unit that analyzes a method for changing the person being evaluated based on the behavioral information; a feedback generation unit that generates feedback information for the person being evaluated based on the analysis results from the first and second analysis units; and an output unit that outputs the generated feedback information.

[0008] An information processing method according to one aspect of the present invention is characterized by a computer performing the following actions: acquiring evaluator information relating to an evaluator who evaluates the actions of a person being evaluated; acquiring behavioral information relating to the actions of the person being evaluated; analyzing the evaluator's behavioral change methods based on the evaluator information; analyzing the person being evaluated's behavioral change methods based on the behavioral information; generating feedback information for the person being evaluated based on the analysis results of analyzing the evaluator's behavioral change methods and the person being evaluated's behavioral change methods; and outputting the generated feedback information.

[0009] A program according to one aspect of the present invention primarily involves causing a computer to perform the following actions: acquire evaluator information relating to an evaluator who evaluates the actions of a person being evaluated; acquire behavioral information relating to the actions of the person being evaluated; analyze the evaluator's behavioral change methods based on the evaluator information; analyze the person being evaluated's behavioral change methods based on the behavioral information; generate feedback information for the person being evaluated based on the analysis results of the analysis of the evaluator's behavioral change methods and the person being evaluated's behavioral change methods; and output the generated feedback information. [Effects of the Invention]

[0010] According to the disclosure technology, it is possible to provide feedback information on the behaviors of the person being evaluated, taking into account their behavioral characteristics. [Brief explanation of the drawing]

[0011] [Figure 1] This figure shows an example of the configuration of the information processing system according to this embodiment. [Figure 2] This figure shows an example of the hardware configuration of the information processing device and terminal device according to this embodiment. [Figure 3] This figure shows an example of the functional block configuration of the information processing device according to this embodiment. [Figure 4] This figure shows an example of an evaluator information database according to this embodiment. [Figure 5] This figure shows an example of a behavioral information database according to this embodiment. [Figure 6] This figure shows an example of a nudge element database according to this embodiment. [Figure 7] This figure shows an example of a graph displayed on the terminal device according to this embodiment. [Figure 8] This figure shows an example of a graph displayed on the terminal device according to this embodiment. [Figure 9] This figure shows an example of a graph displayed on the terminal device according to this embodiment. [Figure 10] This figure shows an example of the functional block configuration of the terminal device according to this embodiment. [Figure 11] This is a flowchart of the processing procedure of the information processing device according to this embodiment. [Modes for carrying out the invention]

[0012] A preferred embodiment of the disclosed technology will be described with reference to the attached drawings. In each drawing, components denoted by the same reference numerals have the same or similar configuration.

[0013] <Information Processing System 1> Hereinafter, the information processing system 1 in the disclosed technology will be described. FIG. 1 is a diagram showing an example of the configuration of the information processing system 1 according to the present embodiment. The information processing system 1 shown in FIG. 1 includes an information processing device 10 and one or more terminal devices 20. The information processing device 10, the terminal device 20, and the external device 30 are connected to each other so as to be communicable via a network N. Note that the external device 30 is not an essential component for configuring the information processing system 1.

[0014] The network N is a network for communication between the information processing device 10, the terminal device 20, and the external device 30. For example, the network N may be any of the Internet, an intranet, a LAN, a mobile communication network, a dedicated line, a packet communication network, a telephone line, a corporate internal network, other communication lines, combinations thereof, and the like. Also, the network N may be wired or wireless.

[0015] <照 Hereinafter, the action performed by the evaluated person according to the present embodiment may be, for example, a presentation performed by the evaluated person. Also, in that case, the evaluated person is the presenter who presents the presentation, and the evaluator is the audience who listens to the presentation. Note that the action performed by the evaluated person is not limited to this, and may be any action such as role-playing in customer service such as the sale of financial products provided by a financial institution, information related to behavior change techniques (BCT: Behavior Change Technique) (hereinafter also referred to as BCT) such as language learning and diet support, etc. Hereinafter, the information related to BCT will also be referred to as BCT information

[0016] Also, in addition to the presenter who presents the presentation, the evaluated person may be a customer service person who is the subject of the action in customer service, an instructor for language learning, etc. Also, the evaluator may be, for example, an actual person having expertise in an arbitrary specialized field. The evaluator may be, for example, a person well-known in a specific field or generally in the world. The evaluator may be a person selected by the evaluated person.

[0017] <Information processing device 10> The information processing device 10 is constituted by a server, a personal computer, or the like. The information processing device 10 is, for example, an information processing device 10 that undertakes part of the functions of information processing provided by the information processing system 1, such as receiving actions performed by the evaluated person and generating feedback information. The information processing device 10 may be constituted using a virtual server, a cloud server, or the like. The information processing device 10 may be called a computer.

[0018] In the present embodiment, the information processing device 10 acquires action information regarding the actions of the evaluated person. For example, the actions of the evaluated person include the mutual communication between the AI avatar and the evaluated person. The information processing device 10 can generate an AI avatar that mimics the evaluator by learning using evaluator information regarding the evaluator who evaluates the actions performed by the evaluated person. The mutual communication between the AI avatar and the evaluated person may be performed via a display or the like of the terminal device 20. For example, an image of the AI avatar or the content of the conversation may be displayed. The information processing device 10 can, for example, output a chat including feedback information to the evaluated person via the AI avatar and support the improvement of the evaluated person's presentation. Note that the mutual communication between the AI avatar and the evaluated person may be performed via a speaker, a microphone, or the like.

[0019] An AI avatar is a virtual entity equipped with AI, and includes virtual characters that mimic evaluators who assess the actions of the person being evaluated. An AI avatar may also be a system or software program that autonomously performs a specific task. Furthermore, an AI avatar may utilize generative AI such as a Large Language Model (LLM) or Large Action Model (LAM) to evaluate the actions of the person being evaluated. The AI ​​avatar can output the generated response information, for example, as text data or audio data. In addition, when outputting response information from the terminal device 20, the response information can be output together with a video or image that makes it appear as if the AI ​​avatar mimicking the evaluator is actually speaking. An AI avatar may be synonymous with a so-called AI agent, etc.

[0020] Furthermore, a generation AI such as a large-scale language model may be realized by the cooperation of an external device 30 having generation AI functions and an information processing device 10, and an AI avatar may be realized by the cooperation of an external device 30 having functions related to the generation and operation of the AI ​​avatar and an information processing device 10. In addition, the external device 30 having generation AI functions and the external device 30 having functions related to the generation and operation of the AI ​​avatar may be the same external device, or they may be separate external devices. <Terminal device 20>

[0021] The terminal device 20 is a device used by the person being evaluated, and is, for example, a mobile phone (including a smartphone), a tablet, or a personal computer. The terminal device 20 provides various information to the information processing device 10 by outputting behavioral information, etc. The person being evaluated can display feedback information, etc., output by the information processing device 10 by operating the terminal device 20. The terminal device 20 may also be called a computer.

[0022] Furthermore, the terminal device 20 may have application programs (apps) installed for using various functions provided by the information processing system 1. These apps may be web browsing software. The apps may cause the terminal device 20 to execute at least a portion of the processing disclosed in the embodiments shown below, within the various functions provided by the information processing system 1. When these apps are executed, the terminal device 20 may access the information processing device 10 to send and receive information used for executing the apps. <External device 30>

[0023] External device 30 is, for example, a computer that operates a large-scale language model that processes input strings of natural language. External device 30 stores a large amount of data used in the large-scale language model and uses the data stored in the database to perform processing associated with the large-scale language model.

[0024] The external device 30 may be configured as a generative AI using a large-scale language model. The generative AI can use an existing generative AI. For example, the external device 30 can send and receive information with the information processing device 10 via the generative AI's API (Application Programming Interface).

[0025] External device 30 includes, for example, services that can be provided on the cloud. External device 30 may be, for example, a generative AI server that provides cloud-based services using a large-scale language model such as OpenAI's ChatGPT. However, the generative AI server is not limited to this, and may be a generative AI server that performs natural language processing using a large-scale language model that provides similar functionality.

[0026] For example, the external device 30 may generate feedback information based on the analysis results of the person being evaluated and the analysis results of the evaluator, which are output from the information processing device 10. The external device 30 may also output the generated feedback information to the information processing device 10.

[0027] <Hardware Configuration> Figure 2 shows an example of the hardware configuration of the information processing device 10 and terminal device 20 according to this embodiment. The information processing device 10 and terminal device 20 include a processor 11 such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit), a storage device 12 such as memory (e.g., RAM (Random Access Memory) or ROM (Read Only Memory)), an HDD (Hard Disk Drive) and / or SSD (Solid State Drive), a communication interface 13 for wired or wireless communication, an input device 14 for receiving input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, touch panel, mouse, camera and / or microphone. The output device 15 is, for example, a display, touch panel and / or speaker. The input device 14 receives input of various information from the user. The output device 15 may also include a display unit that displays various information to the user. The external device 30 may also have the same hardware configuration as described above.

[0028] The hardware configuration described above is merely an example. The information processing device 10 and terminal device 20 within the information processing system 1 may omit some of the hardware shown in Figure 2, or may include hardware not shown in Figure 2. Furthermore, the hardware shown in Figure 2 may be composed of one or more devices. Also, if the information processing device 10 is composed of multiple devices, each device may include at least some of this hardware, and the same applies to the terminal device 20.

[0029] <Functional Block Configuration> (Information processing device 10) Figure 3 shows an example of the functional block configuration of the information processing device 10 according to this embodiment. The information processing device 10 includes a storage unit 100, a first acquisition unit 101, a second acquisition unit 102, a first analysis unit 103, a second analysis unit 104, a feedback generation unit 105, and an output unit 106. The information processing device 10 may further include an AI avatar generation unit 107, a graph generation unit 108, and a similarity calculation unit 109.

[0030] The memory unit 100 can be implemented using the memory device 12 provided by the information processing device 10. Furthermore, the first acquisition unit 101, the second acquisition unit 102, the first analysis unit 103, the second analysis unit 104, the feedback generation unit 105, the output unit 106, the AI ​​avatar generation unit 107, the graph generation unit 108, and the similarity calculation unit 109 can be implemented by the processor 11 of the information processing device 10 executing a program stored in the memory device 12.

[0031] Furthermore, the program may be stored in a storage medium. The storage medium on which the program is stored may be a non-transitory computer-readable medium. The non-transitory storage medium is not particularly limited, but may be, for example, a USB (Universal Serial Bus) memory or a CD-ROM (Compact Disc Read-Only Memory).

[0032] The storage unit 100 stores the data necessary for the information processing device 10 to perform information processing. The storage unit 100 includes an evaluator information DB 100a, an action information DB 100b, a nudge element DB 100c, and an extraction model 100d. It is possible to add data items to each DB as needed.

[0033] Figure 4 shows an example of the evaluator information DB 100a according to this embodiment. The evaluator information DB 100a shown in Figure 4 manages various information about evaluators. The evaluator information DB 100a may store evaluator IDs, evaluator information IDs, and evaluator information in association with each other.

[0034] The evaluator ID is information that identifies the evaluator. The evaluator information ID is information that identifies the evaluator's information. The evaluator information stores information about the evaluator.

[0035] Evaluator information refers to information about the evaluator, and may include profile information such as name, gender, age, address, date of birth, work history, qualifications, special skills, personality, and family structure. Evaluator information may also include information on the results of questionnaires answered by the evaluator. The questionnaire content may include questions about behavioral change methods such as the evaluator's values ​​and characteristics, or it may include questions about the evaluator's daily life. Evaluator information may also include books written by the evaluator, social networking services (SNS) operated by the evaluator, blogs, and videos in which the evaluator is interviewed. For videos, a maximum length limit for storage (e.g., up to 5 minutes) may be set in advance. Information regarding the date and time each evaluator information was acquired may be associated with and stored in the evaluator information.

[0036] Figure 5 shows an example of the behavioral information DB 100b according to this embodiment. The behavioral information DB 100b shown in Figure 5 manages various types of information related to behavioral information. The behavioral information DB 100b may store the evaluated person ID, the behavioral information ID, and the behavioral information in association with each other.

[0037] The Evaluated Person ID is information that identifies the person being evaluated. The Behavioral Information ID is information that identifies the behavioral information. The behavioral information stores information about the person being evaluated's behavior.

[0038] Behavioral information includes information about the behavior of the person being evaluated. For example, behavioral information may include video of the person being evaluated's movements captured using a camera, audio of the person being evaluated's speech captured using a microphone, text data such as audio converted to text, and text data such as presentation materials. Behavioral information can be any data for which feedback is requested. For example, it may include sales data provided by a sales representative to a customer, learning support data provided by an instructor to a learner, or training data performed by an athlete. For videos, an upper limit on the length to be stored (e.g., up to 5 minutes) may be set in advance.

[0039] Behavioral information may include information about the mutual communication between the AI ​​avatar and the person being evaluated. For example, behavioral information such as a video of the person being evaluated giving a presentation may be input to the AI ​​avatar. Based on the input behavioral information, the AI ​​avatar outputs response information that is predicted to be the evaluator's reaction. The information regarding the mutual communication between the AI ​​avatar and the person being evaluated described above may also be stored as behavioral information.

[0040] Furthermore, behavioral information may include, for example, profile information of the person being evaluated, such as name, gender, age, address, date of birth, work history, qualifications, special skills, personality, and family structure. Behavioral information may also include information on the results of questionnaires answered by the person being evaluated. The questionnaire content may include, for example, questions related to behavioral change methods such as the person being evaluated's values ​​and characteristics, or it may include questions related to the person being evaluated's daily life.

[0041] Figure 6 shows an example of a nudge element DB100c according to this embodiment. The nudge element DB100c shown in Figure 6 manages nudge elements associated with BCTs. The nudge element DB100c may store a nudge element ID, a BCT, and a nudge element in association with each other. Note that the BCT, nudge elements, etc., stored in the nudge element DB100c may be changed as appropriate.

[0042] The nudge element ID is information that identifies a BCT. A BCT is a technique or method that acts on (influences) behavioral change. For example, according to BCTTv1 (Michie S, Richardson M, Johnstonm, Et Al.: The Behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Ann Behav Med 2013; 46: 81~95.), 93 BCTs in 16 groups are defined. Note that the definition of a BCT is not limited to BCTTv1; it may be defined in any way as long as it comprehensively covers methods of behavioral change.

[0043] Furthermore, the BCTs in this embodiment include groups containing one or more BCTs (hereinafter referred to as "BCT groups") (for example, the BCTTv1 group), and BCT groups and BCTs may be interchangeable. In the example shown in Figure 6, 12 BCTs are shown: "Goals and planning," "Feedback & Monitoring," "Social support," "Shaping knowledge," "Antecedents," "Comparison of behavior," "Reward & Threat," "Comparison of outcomes," "Associations," "Identity," "Scheduled consequences," and "Self-belief." Note that each BCT shown in Figure 6 may be a BCT group containing one or more BCTs, and nudge elements may be associated with each BCT within that BCT group.

[0044] Furthermore, a nudge element is one or more elements associated with a BCT. A nudge element may be, for example, the text of a message regarding feedback using a BCT. A nudge element may also be configured to store the text of a message to which the BCT is applied, or it may be configured to store information about a person who possesses the behavioral characteristics of the BCT. A nudge element may also be configured to store nudge elements that possess the behavioral characteristics of the BCT in association with the BCT.

[0045] Extraction model 100d is a model that extracts information about BCT contained in the evaluator information, which is the input data. Extraction model 100d may also be a model that extracts information about the evaluator's behavioral change methods based on the evaluator information. Alternatively, extraction model 100d is a model that extracts information about BCT contained in the behavioral information, which is the input data. Extraction model 100d may also be a model that extracts information about the evaluated person's behavioral change methods based on the behavioral information. The evaluator information and behavioral information, which are the input data, may be in data formats such as text, images, videos, and audio. Extraction model 100d may also be a model that extracts biases regarding behavioral change methods.

[0046] The extraction model 100d may be generated based on machine learning using evaluator information and BCT information as training data, or it may be generated based on machine learning using behavioral information and BCT information as training data. Alternatively, the extraction model 100d may be generated by training it using data to which BCT annotation information has been added. By using the extraction model 100d, BCT information can be extracted from both behavioral information and evaluator information. The extraction model 100d may be a single model 100d that includes a model for extracting BCT information contained in evaluator information and a model for extracting BCT information contained in behavioral information. Alternatively, the extraction model 100d may use separate extraction models 100d corresponding to the model for extracting BCT information contained in evaluator information and the model for extracting BCT information contained in behavioral information.

[0047] Furthermore, each BCT has components, and the extraction model 100d may extract the degree of each BCT contained in the evaluator information and behavioral information, respectively, as component values. Alternatively, the extraction model 100d may extract the sum of the component values ​​of each BCT belonging to the same BCT group as the component value of the BCT group. Return to Figure 3 and continue the explanation.

[0048] The first acquisition unit 101 acquires evaluator information about the evaluator who evaluates the actions performed by the person being evaluated. Evaluator information may include, for example, the evaluator's lifestyle, values, hobbies, and other unique information, and may also include information that contains many characteristics of the evaluator. The first acquisition unit 101 may also acquire evaluator information based on the questionnaire results, for example, by acquiring the questionnaire results answered by the evaluator. Alternatively, the person being evaluated may operate the terminal device 20 to select an evaluator, and the first acquisition unit 101 may acquire the information of the selected evaluator (for example, expert A). Subsequently, the first acquisition unit 101 acquires the evaluator information of expert A, who was selected by the person being evaluated.

[0049] Furthermore, the first acquisition unit 101 may acquire, for example, the evaluator's attribute information (age, gender, location, field of expertise, hobbies, etc.), SNS information posted by the evaluator, blogs, books, interview videos in which the evaluator appeared, etc., as evaluator information. The first acquisition unit 101 may also acquire evaluator information such as SNS information as needed and update the evaluator information. In addition, the first acquisition unit 101 may automatically acquire evaluator information from the internet or other sources.

[0050] The second data acquisition unit 102 acquires behavioral information about the actions of the person being evaluated. The actions of the person being evaluated include, for example, presentations given by the person being evaluated. The second data acquisition unit 102 may acquire, as behavioral information, for example, text (presentation materials, etc.) related to the presentation given by the person being evaluated, videos and still images of the presentation being given, audio of the presentation being given, and information about the application used for the presentation.

[0051] The first analysis unit 103 analyzes the methods of behavioral change of evaluators based on evaluator information. For example, the first analysis unit 103 inputs the evaluator information acquired by the first acquisition unit 101 into the extraction model 100d and extracts one or more BCT information. Hereinafter, the BCT information extracted from the evaluator information will also be referred to as the evaluator analysis results.

[0052] The second analysis unit 104 analyzes the behavioral change methods of the person being evaluated based on the behavioral information. For example, the second analysis unit 104 inputs the behavioral information acquired by the second acquisition unit 102 into the extraction model 100d and extracts one or more BCT information. Hereinafter, the BCT information extracted from the behavioral information will also be referred to as the analysis results of the person being evaluated.

[0053] Furthermore, the BCT information extracted by the first analysis unit 103 and the second analysis unit 104 may be a component table showing fixed values ​​based on the average of the BCT components contained in the evaluator information and behavioral information, respectively, or it may show the distribution of the BCT components contained in the evaluator information and behavioral information, respectively.

[0054] Furthermore, the BCT information extracted by the first analysis unit 103 and the second analysis unit 104 may include information regarding the ratio of the amount of information of the BCT components to the amount of information of the behavioral information and the evaluator information, respectively. In addition, the BCT information may include information regarding the ratio of the amount of information of a specific BCT component to the amount of information corresponding to the BCT components of the behavioral information and the evaluator information, respectively.

[0055] Furthermore, the first analysis unit 103 and the second analysis unit 104 may analyze audio information, including speech duration, content, volume, and pitch, and extract BCT information. Alternatively, the first analysis unit 103 and the second analysis unit 104 may analyze video information, including speech duration, content, volume, and pitch, and extract BCT information. Furthermore, the first analysis unit 103 and the second analysis unit 104 may analyze each piece of information from a predetermined perspective and extract BCT information.

[0056] The feedback generation unit 105 generates feedback information on the evaluated person's behavior based on the analysis results from the first analysis unit 103 and the second analysis unit 104. The feedback generation unit 105 may also compare the evaluator's analysis results with the evaluated person's analysis results and generate feedback information on the evaluated person's behavior based on the comparison results. The feedback generation unit 105 may generate feedback information that includes nudge elements associated with BCT based on the comparison results. The feedback generation unit 105 may also generate feedback information on the strengths and weaknesses of the evaluated person's behavior based on the comparison results.

[0057] Furthermore, the feedback generation unit 105 may compare the analysis results of the person being evaluated with the analysis results of the evaluator and control the number or amount of nudge elements to be included in the feedback information based on the magnitude of the difference in each BCT. For example, the feedback generation unit 105 may increase the number or amount of nudge elements associated with a BCT (e.g., goals and plans) included in the feedback information the larger the difference between the analysis results of the person being evaluated and the evaluator (e.g., goals and plans). On the other hand, the feedback generation unit 105 may decrease the number or amount of nudge elements associated with a BCT (e.g., social support) included in the feedback information the smaller the difference between the analysis results of the person being evaluated and the evaluator (e.g., social support).

[0058] Feedback information may include, for example, evaluation items for feedback and types of feedback. The types of feedback may include, for example, positive feedback and negative feedback. The feedback generation unit 105 may determine one or more evaluation items based on the categories of behavioral information and generate feedback for each evaluation item.

[0059] The feedback generation unit 105 may generate feedback information by color-coding each evaluation item. The feedback generation unit 105 may also generate feedback information by color-coding each extracted BCT.

[0060] For example, in the case of the behavioral information category "Presentation," the feedback generation unit 105 may determine "Structure," "Content," "Presentation," and "Materials" as evaluation items and generate feedback for each evaluation item. For "Structure," it evaluates whether the presentation is in line with the purpose and theme, and whether it has a logical and consistent flow. For "Content," it evaluates whether the supporting literature and data are presented appropriately, and whether the information is convincing to the audience. For "Presentation," it evaluates whether the volume and speaking speed are appropriate, and whether eye contact is maintained appropriately. For "Materials," it evaluates whether the materials are of a readable size, whether they are in the form of charts and graphs, whether the writing is appropriate, and whether the information is easy to grasp intuitively.

[0061] In another example, for the behavioral information category "insurance sales role-playing," the feedback generation unit 105 can determine "listening skills," "product description skills," "proposal skills," and "trouble-solving skills" as evaluation items and generate feedback for each evaluation item. The feedback generation unit 105 may generate one piece of feedback for each evaluation item, or it may generate multiple pieces of feedback for each evaluation item. For example, the feedback generation unit 105 may generate both positive and negative feedback for a single evaluation item. For instance, for an evaluation item related to posture, it may generate positive feedback such as "The person had good posture with a straight back" and negative feedback such as "The person touched their face frequently with their hands, giving the impression of being restless."

[0062] The feedback generation unit 105 may, for example, cooperate with an external device 30 to generate feedback information. Specifically, the feedback generation unit 105 generates generation request information that incorporates the analysis results of the person being evaluated and the evaluator into pre-set generation request information (so to speak, a prompt). The feedback generation unit 105 may then use the generated generation request information to obtain feedback information generated by the external device 30 and output the feedback information.

[0063] The output unit 106 outputs the generated feedback information. In addition to the feedback information, the output unit 106 may also output the facial expressions and emotions of the AI ​​avatar. For example, the output unit 106 may output an image of the appearance of the AI ​​avatar. This makes it possible to output feedback information that includes visual and auditory information about the evaluator.

[0064] With the above configuration, the information processing device 10 can provide feedback information on the actions taken by the person being evaluated, taking into account the behavioral characteristics of the person being evaluated. Furthermore, the person being evaluated can recognize biases and distortions in their own actions by receiving feedback information on their own actions, which is generated based on the results of analyzing the evaluator's behavioral change methods based on the evaluator's information, and based on the results of analyzing the person being evaluated's behavioral change methods based on their own behavioral information.

[0065] The AI ​​avatar generation unit 107 generates an AI avatar of the evaluator by learning using the evaluator information. For example, the AI ​​avatar generation unit 107 may use the acquired evaluator information as training data to generate an AI avatar that simulates the evaluator's thinking and behavioral patterns through machine learning processing. Alternatively, the AI ​​avatar generation unit 107 may generate an AI avatar that can predict the evaluator's reaction to behavioral information by learning the evaluator information. The AI ​​avatar generation unit 107 may also generate an AI avatar of the evaluator by learning using the evaluator information acquired by the first acquisition unit.

[0066] Furthermore, the AI ​​avatar may be generated to output a response as response information that is predicted to be the evaluator's answer when question data is provided. The AI ​​avatar generation unit 107 may also automatically perform machine learning processing when new evaluator information is acquired. This allows the evaluator's AI avatar to simulate the evaluator's thinking and behavior patterns based on the evaluator's latest information. The generated AI avatar may also be stored in the memory unit. The AI ​​avatar generation unit 107 may, for example, cooperate with an external device 30, which is an AI avatar generation system, and acquire the generated AI avatar from the AI ​​avatar generation system. It may also acquire response information output by the AI ​​avatar from the AI ​​avatar generation system.

[0067] The actions of the person being evaluated may include mutual communication between the AI ​​avatar and the person being evaluated. The second acquisition unit 102 may acquire behavioral information about the person being evaluated, including information about the mutual communication between the AI ​​avatar and the person being evaluated. The actions of the person being evaluated may include, for example, a presentation given by the person being evaluated. The second acquisition unit 102 may acquire, as behavioral information, for example, text (presentation materials, etc.) related to the presentation given by the person being evaluated, videos and still images of the presentation being given, audio of the presentation being given, and information about the application used for the presentation.

[0068] Furthermore, the AI ​​avatar generation unit 107 may input behavioral information such as videos, audio, and presentation materials used by the person being evaluated, which have been acquired by the second acquisition unit 102, into the AI ​​avatar. Based on the input behavioral information, the AI ​​avatar may generate and output response information that is predicted to be the evaluator's own reaction. The second acquisition unit 102 acquires the outputted response information as mutual communication information between the AI ​​avatar and the person being evaluated. Since the AI ​​avatar is generated by machine learning processing so as to be able to simulate the evaluator's thinking and behavioral patterns, the person being evaluated can obtain a simulated response equivalent to the response obtained from the evaluator themselves.

[0069] The output response information may be text data indicating a question generated based on the input behavior information. The AI ​​avatar generation unit 107 may also input question data for the AI ​​avatar received from the person being evaluated. The AI ​​avatar may output a response information that is predicted to be the answer the evaluator will give to the question data, and this mutual communication information may be acquired by the second acquisition unit 102.

[0070] The AI ​​avatar may, for example, generate a question about the content included in the presentation materials of the person being evaluated, if the person makes a statement during the presentation that differs from the content, or if the person fails to present the content in question, and output the response information. Based on the response information, the person being evaluated may correct their statements or add more detailed explanations about the content. The AI ​​avatar may output a response information indicating whether or not it is satisfied with the corrections made by the person being evaluated. This mutual communication information may be acquired by the second acquisition unit 102. The person being evaluated may change their behavior based on the response information generated by the AI ​​avatar. The second acquisition unit 102 may acquire the information about the changed behavior.

[0071] The AI ​​avatar may output response information such as impressions of the video, simplified reaction information such as "good" and "bad," current evaluation (e.g., how many out of 5), and comments. The second acquisition unit 102 may acquire the output information as mutual communication information. "Good" and "bad" may be sent when the AI ​​avatar has positive or negative feelings towards the video. They may also be output when an AI avatar modeled after an evaluator wants to show interest or concern to the person being evaluated.

[0072] Through the above processing, it is possible to provide feedback information on the behavior of the person being evaluated, including the interactive communication between the AI ​​avatar and the person being evaluated, while taking into account the behavioral characteristics of the person being evaluated. By receiving feedback generated using an AI avatar modeled after the evaluator, the person being evaluated can recognize biases and distortions in their own behavior and gain an opportunity to make changes.

[0073] Furthermore, by utilizing AI technology, the actions performed by those being evaluated and the evaluation process for evaluators can be significantly streamlined, improving convenience and satisfaction for those being evaluated. For example, those being evaluated can receive evaluations of their actions using an AI avatar modeled after an evaluator, regardless of time or location, thus reducing their burden and improving convenience. In addition, those being evaluated can flexibly obtain feedback information from the AI ​​avatar modeled after an evaluator according to their own progress, enabling them to efficiently improve their actions. In other words, by utilizing an AI avatar modeled after an evaluator, the burden of providing feedback to evaluators can be significantly reduced, thus alleviating the burden on evaluators as well.

[0074] The graph generation unit 108 generates a graph based on the analysis results from the first analysis unit 103 and the second analysis unit 104. The graph generation unit 108 may also generate a graph based on the analysis results of the person being evaluated and the analysis results of the evaluator. For example, the graph generation unit 108 may generate a graph in the form of a radar chart based on the analysis results of the person being evaluated and the analysis results of the evaluator.

[0075] The output unit 106 may output the generated graph. Figures 7 and 8 show examples of graphs displayed on the terminal device 20 according to this embodiment. Figure 7 is a radar chart showing the components of 12 BCTs based on the analysis results of the person being evaluated, and Figure 8 is a radar chart showing the components of 12 BCTs based on the analysis results of the evaluator. Note that each BCT may represent the components of a BCT group that includes one or more BCTs under it, or it may be the sum of the components of such one or more BCTs. This allows the person being evaluated to perceive a larger amount of information by using graphs, in addition to text and audio information.

[0076] The graph generation unit 108 may generate graphs that reflect the analysis results of the person being evaluated and the analysis results of the evaluator on one or more radar charts. Figure 9 shows an example of a graph displayed on the terminal device 20 according to this embodiment. This graph reflects the behavioral analysis results of the person being evaluated and the analysis results of the evaluator on a single radar chart showing the components of 12 BCTs according to this embodiment. As shown in Figure 9, the behavioral analysis results of the person being evaluated and the analysis results of the evaluator can be easily compared, and the similarities, differences, outliers, balance of BCT composition, and characteristics of each analysis result can be clearly expressed. In addition, multiple BCTs can be displayed in a highly visible manner, and the magnitude of the component amounts of each BCT can be grasped instantly.

[0077] The similarity calculation unit 109 calculates the similarity score based on the analysis results of the person being evaluated and the analysis results of the evaluator. The similarity calculation unit 109 may, for example, calculate the similarity score based on the feature quantities of the analysis results of the person being evaluated and the evaluator. The similarity score may be calculated using a statistical similarity method, or it may be calculated using an approximate method such as clustering.

[0078] The feedback generation unit 105 may generate feedback information regarding a recognition discrepancy if the similarity calculated by the similarity calculation unit 109 is below a predetermined threshold.

[0079] Through the above process, if the similarity between the analysis results of the person being evaluated and the analysis results of the evaluator is below a predetermined threshold, feedback information regarding a discrepancy in perception can be generated. This allows the person being evaluated to understand that their behavior and the evaluator's BCT are dissimilar and that a discrepancy in perception has occurred.

[0080] The feedback generation unit 105 may generate feedback information for the person being evaluated based on the discrepancies between the person being evaluated's analysis results and the evaluator's analysis results.

[0081] The feedback generation unit 105 may, for example, compare the analysis results of the person being evaluated with the analysis results of the evaluator and identify the BCTs to include in the feedback information for the person being evaluated based on the difference between the two analysis results. Specifically, by comparing the analysis results of the person being evaluated with the analysis results of the evaluator, at least a portion of the BCTs in which the components of the person being evaluated are lacking may be identified as the BCTs to include in the feedback information. The feedback generation unit 105 may generate feedback information that includes one or more nudge elements associated with the identified BCTs.

[0082] The feedback generation unit 105 may, for example, identify the BCT (Brain-Challenge Trick) where the difference between the evaluated person's analysis results and the evaluator's analysis results is greatest as the BCT causing the discrepancy in perception.

[0083] Furthermore, the feedback generation unit 105 may generate feedback information related to resolving discrepancies in perception between the person being evaluated and the evaluator. The feedback generation unit 105 may generate feedback information that includes one or more nudge elements associated with the BCT that is the cause of the identified discrepancy in perception. The feedback generation unit 105 may also identify BCTs that are missing from the person being evaluated's analysis results and generate feedback information that complements them.

[0084] Through the above process, the person being evaluated can receive feedback information based on the discrepancies between their own analysis results and the evaluator's analysis results, enabling them to make targeted improvements to their behavior. As a result, the constraints on the person being evaluated's skill development can be removed.

[0085] The feedback generation unit 105 may generate feedback information regarding the difference between the evaluation results of the person being evaluated and the evaluation results of the evaluator. For example, the feedback generation unit 105 may generate feedback information that includes nudge elements associated with each of the BCTs where a difference exists between the evaluation results of the person being evaluated and the evaluation results of the evaluator. This makes it possible to generate feedback information that applies a more comprehensive BCT from a behavioral science perspective.

[0086] The feedback generation unit 105 may generate feedback information that includes nudge elements associated with each of the BCTs. The feedback generation unit 105 may also generate feedback information so that the analysis results of the behavioral information approach the analysis results of the evaluator.

[0087] Furthermore, in order for the analysis results of the behavioral information to approach the analysis results of the evaluator, the feedback generation unit 105 may, for example, have the person being evaluated select a specific BCT with high priority to emphasize when generating feedback information. Alternatively, the decision on which BCT to emphasize when generating feedback information may be based on the magnitude of the difference between the BCT component of the behavioral information analysis results and the BCT component of the evaluator's analysis results, or it may be based on environmental constraints related to the implementation system.

[0088] Furthermore, the feedback generation unit 105 may generate feedback information regarding BCT that is considered effective in identifying factors that promote and / or inhibit the behavior of the person being evaluated and in encouraging improvement of behavioral information. For example, the feedback generation unit 105 may have the person being evaluated choose whether to aim for a specialized type that matches the evaluator's behavioral change method, or a balanced type that supplements areas that are lacking, and generate feedback information based on the selected content.

[0089] For example, if the person being evaluated chooses to aim for specialization, the feedback generation unit 105 may generate feedback information that increases the proportion / purity of specific BCTs to match the evaluator's behavioral change method. Alternatively, if the person chooses to aim for a balanced approach, the unit may generate feedback information that supplements any missing BCTs in the behavioral information.

[0090] Through the above process, the person being evaluated can obtain specific and actionable improvement plans based on the feedback information, enabling them to formulate concrete improvement measures from a behavioral science perspective. Furthermore, the strengths and weaknesses of the person being evaluated become clear from a behavioral science perspective, making it easier to predict what effects the improvements will have. In addition, the improvement of the person being evaluated's metacognition can contribute to their self-perception and perception of others.

[0091] (Terminal device 20) Figure 10 shows an example of the functional block configuration of the terminal device 20 according to this embodiment. The terminal device 20 includes a storage unit 200, a UI (User Interface) unit 201, and a control unit 202. The storage unit 200 can be implemented using a storage device 12 provided by the terminal device 20. The UI unit 201 and the control unit 202 can be implemented by the processor 11 of the terminal device 20 executing a program stored in the storage device 12. The program can be stored in a storage medium. The storage medium on which the program is stored may be a computer-readable non-temporary storage medium. The non-temporary storage medium is not particularly limited, but may be, for example, a USB memory or a CD-ROM.

[0092] The memory unit 200 stores data and other information necessary for the control unit 202 to perform information processing.

[0093] The UI unit 201 has the function of receiving various inputs from the person being evaluated and displaying various screens on the display. In addition, the UI unit 201 may display feedback information or an AI avatar on the display of the terminal device 20 in accordance with the instructions of the information processing device 10.

[0094] The control unit 202, in cooperation with the information processing device 10, provides various functions necessary for executing information processing. For example, the control unit 202 may receive feedback information from the information processing device 10 and notify the person being evaluated of the feedback information by displaying the received feedback information on the output unit 106.

[0095] Furthermore, the control unit 202 may collect or acquire various evaluator information. The control unit 202 may also control the system to transmit the acquired evaluator information to the information processing device 10.

[0096] Regarding the functional block configuration described above, it is also possible to configure the terminal device 20 to include all or part of the following components included in the information processing device 10: the storage unit 100, the first acquisition unit 101, the second acquisition unit 102, the first analysis unit 103, the second analysis unit 104, the feedback generation unit 105, the output unit 106, the AI ​​avatar generation unit 107, the graph generation unit 108, and the similarity calculation unit 109. In other words, the various processes according to this embodiment may be executed by the processor of the information processing device 10, by the processor of the terminal device 20, or by the processors of the information processing device 10 and the terminal device 20 working together.

[0097] <System Operation>

[0098] Next, the operation of the information processing device 10 according to this embodiment will be described. Figure 11 is a flowchart of the processing procedure of the information processing device 10 according to this embodiment. In Figure 11, a presentation is shown as an example of an action performed by the person being evaluated. In this embodiment, it is assumed that various data are stored in the storage unit 100 before the processing in Figure 11 begins.

[0099] In step S101, the first acquisition unit 101 acquires evaluator information relating to the evaluator who evaluates the actions performed by the person being evaluated. The first acquisition unit 101 may also acquire evaluator information such as SNS information, blogs, books, and interview videos in which the evaluator appeared.

[0100] In step S102, the AI ​​avatar generation unit 107 generates an AI avatar of the evaluator by learning using the evaluator information. The AI ​​avatar generation unit 107 may also use the acquired evaluator information as training data to generate an AI avatar that simulates the evaluator's thinking and behavioral patterns through machine learning processing.

[0101] In step S103, the second acquisition unit 102 acquires behavioral information about the person being evaluated, including information on the mutual communication between the person being evaluated and their AI avatar. The AI ​​avatar generation unit 107 may input the behavioral information acquired by the second acquisition unit 102, such as videos, audio, and materials used in the presentation, into the AI ​​avatar. Based on the input behavioral information, the AI ​​avatar may generate and output response information predicted to be the evaluator's reaction. The second acquisition unit 102 may acquire the outputted response information as information on the mutual communication between the AI ​​avatar and the person being evaluated.

[0102] In step S104, the first analysis unit 103 analyzes the evaluator behavior change methods based on the evaluator information acquired by the first acquisition unit 101. The first analysis unit 103 may also input the evaluator information into the extraction model 100d and extract the BCT information contained in the evaluator information.

[0103] In step S105, the second analysis unit 104 analyzes the behavioral change methods of the person being evaluated based on the behavioral information acquired by the second acquisition unit 102. The second analysis unit 104 may also input the behavioral information into the extraction model 100d and extract the BCT information contained in the behavioral information.

[0104] In step S106, the feedback generation unit 105 generates feedback information on the evaluated person's behavior based on the analysis results from the first analysis unit 103 and the second analysis unit 104. The feedback generation unit 105 may also compare the evaluated person's analysis results with the evaluator's analysis results and generate feedback information on the evaluated person's behavior based on the comparison results. The feedback generation unit 105 may also generate feedback regarding the BCT information included in the evaluator information extracted by the first analysis unit 103 and the BCT information included in the behavior information extracted by the second analysis unit 104. Furthermore, the graph generation unit 108 may generate a graph, for example in the form of a radar chart, based on the evaluated person's analysis results and the evaluator's analysis results.

[0105] In step S107, the output unit 106 outputs the generated feedback information. The output unit 106 may output the generated feedback information as text information or audio information. Alternatively, the output unit 106 may output the graph generated by the graph generation unit 108.

[0106] Through the above processing, it is possible to provide feedback information on the actions taken by the person being evaluated, taking into account their behavioral characteristics.

[0107] Although embodiments of the present invention have been described above, these embodiments or examples are provided to facilitate understanding of the present invention and are not intended to limit it. The present invention can be modified or improved without departing from its spirit, and equivalents thereof are also included. Furthermore, the present invention can form various disclosures by appropriately combining the multiple components disclosed in the above embodiments or examples. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components may be appropriately combined in different embodiments. [Explanation of Symbols]

[0108] 1...Information processing system, 10...Information processing device, 11...Processor, 12...Storage device, 13...Communication interface, 14...Input device, 15...Output device, 20...Terminal device, 30...External device, 100...Storage unit, 100a...Evaluator information DB, 100b...Behavioral information DB, 100c...Nudge element DB, 100d...Extraction model, 101...First acquisition unit, 102...Second acquisition unit, 103...First analysis unit, 104...Second analysis unit, 105...Feedback generation unit, 106...Output unit, 107...AI avatar generation unit, 108...Graph generation unit, 109...Similarity calculation unit, 200...Storage unit, 201...UI unit, 202...Control unit

Claims

1. A first acquisition unit acquires evaluator information about evaluators who evaluate the actions performed by the person being evaluated, A second acquisition unit acquires behavioral information regarding the actions of the person being evaluated, Based on the aforementioned evaluator information, the first analysis unit analyzes the evaluator's method of behavioral change, A second analysis unit analyzes the behavioral change methods of the person being evaluated based on the aforementioned behavioral information, A feedback generation unit generates feedback information regarding the behavior of the person being evaluated based on the analysis results from the first analysis unit and the second analysis unit, The system includes an output unit that outputs the generated feedback information, Information processing device.

2. The system further comprises an AI avatar generation unit that generates an AI avatar of the evaluator by learning using the evaluator information. The aforementioned actions include mutual communication between the AI ​​avatar and the person being evaluated. The information processing apparatus according to claim 1.

3. The system further comprises a graph generation unit that generates a graph based on the analysis results from the first analysis unit and the second analysis unit, The output unit further outputs the generated graph. The information processing apparatus according to claim 1.

4. The graph generation unit generates the graph so as to reflect the analysis results of the person being evaluated and the analysis results of the evaluator on one or more radar charts. The information processing apparatus according to claim 3.

5. The system further comprises a similarity calculation unit that calculates similarity based on the analysis results of the person being evaluated and the analysis results of the evaluator, The feedback generation unit generates feedback information regarding a recognition discrepancy if the calculated similarity is below a predetermined threshold. The information processing apparatus according to claim 1.

6. The feedback generation unit provides feedback to the person being evaluated based on the discrepancies between the person being evaluated's analysis results and the evaluator's analysis results. The information processing apparatus according to claim 1.

7. The feedback generation unit generates feedback information relating to the difference between the analysis results of the person being evaluated and the analysis results of the evaluator. The information processing apparatus according to claim 1.

8. The feedback generation unit provides feedback on matters related to resolving discrepancies in understanding between the person being evaluated and the evaluator. The information processing apparatus according to claim 1.

9. Computers Obtaining evaluator information about the evaluators who evaluate the actions performed by the person being evaluated, To obtain behavioral information regarding the actions of the person being evaluated, Based on the aforementioned evaluator information, the method for changing the evaluator's behavior will be analyzed, Based on the aforementioned behavioral information, the method of behavioral change of the person being evaluated will be analyzed, Based on the analysis results in analyzing the behavioral change methods of the evaluator and the behavioral change methods of the person being evaluated, feedback information regarding the behavior of the person being evaluated is generated. Outputting the generated feedback information, An information processing method that performs the following.

10. On the computer, Obtaining evaluator information about the evaluators who evaluate the actions performed by the person being evaluated, To obtain behavioral information regarding the actions of the person being evaluated, Based on the aforementioned evaluator information, the method for changing the evaluator's behavior will be analyzed, Based on the aforementioned behavioral information, the method of behavioral change of the person being evaluated will be analyzed, Based on the analysis results in analyzing the behavioral change methods of the evaluator and the behavioral change methods of the person being evaluated, feedback information regarding the behavior of the person being evaluated is generated. Outputting the generated feedback information, A program that executes the command.

Images