Information processing method, program, and information processing device

Abstract

[Problem] To prompt a user to correct an answer. [Solution] This information processing method executed by a processor includes: acquiring a user's answer information; acquiring the user's sensing data; calculating the reliability of the answer information on the basis of the user's sensing data at the time the answer information was acquired; determining the importance of the answer information on the basis of the contents of a dialog between the user and another user; and performing display control to prompt the user to correct the answer information according to the importance and the reliability of the answer information.

Description

Information processing method, program, and information processing device

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

[0002] Traditionally, in insurance consultations for life insurance, medical insurance, etc., sales representatives engage in dialogue with customers, and based on the information obtained from the customer during the dialogue, they formulate and present a life plan or propose appropriate insurance products. In various interviews, not only for insurance consultations but also for loan consultations such as education loans and mortgages, or consultations on financial products, sales representatives listen to the customer while talking to gather information necessary for proposing products, services, etc. In such interviews, it is desirable for customers to answer the sales representative's questions honestly, but it is also possible that customers may give answers that are inaccurate due to embarrassment, vanity, or impulsive thoughts.

[0003] Patent Literature 1 proposes a technology for determining the reliability of a patient's answers in a medical interview conducted at a medical institution. By automatically determining the reliability of the patient's answers, the burden on doctors can be reduced.

[0004] JP 2014-089650 A

[0005] However, the technology described in Patent Document 1 presents the reliability determination result to the doctor, and does not obtain a new correct answer from the patient even if the reliability is low.

[0006] Therefore, the present disclosure proposes an information processing method, a program, and an information processing device that can prompt the user to correct their answer.

[0007] According to the present disclosure, an information processing method is provided, including: a processor acquiring answer information of a user; acquiring sensing data of the user; calculating the reliability of the answer information based on the sensing data of the user at the time the answer information was acquired; determining the importance of the answer information based on the content of a dialogue between the user and another user; and controlling a display to prompt the user to correct the answer information according to the importance and reliability of the answer information.

[0008] Furthermore, according to the present disclosure, a program is provided that causes a computer to function as a control unit that performs the following processes: acquiring a user's response information; acquiring sensing data of the user; calculating the reliability of the response information based on the user's sensing data at the time the response information was acquired; determining the importance of the response information based on the content of a dialogue between the user and another user; and controlling the display to prompt the user to correct the response information according to the importance and reliability of the response information.

[0009] Furthermore, according to the present disclosure, there is provided an information processing device that includes a control unit that performs the following processes: acquiring a user's response information; acquiring sensing data of the user; calculating the reliability of the response information based on the user's sensing data at the time the response information was acquired; determining the importance of the response information based on the content of a dialogue between the user and another user; and controlling the display to prompt the user to correct the response information according to the importance and reliability of the response information.

[0010] FIG. 1 is a configuration diagram of an information processing system 1 corresponding to an embodiment of the present disclosure. FIG. 2 is a block diagram showing an example of the configuration of a server 20 according to the embodiment. FIG. 3 is a block diagram showing an example of the configuration of an information processing terminal 100 according to an embodiment of the present disclosure. FIG. 4 is a flowchart showing an example of the flow of operation processing according to the embodiment. FIG. 5 is a diagram showing an example of a customer screen and a sales representative screen according to the embodiment. FIG. 6 is a diagram showing another example of a customer screen and a sales representative screen according to the embodiment. FIG. 7 is a diagram showing another example of a customer screen and a sales representative screen according to the embodiment. FIG. 8 is a diagram showing another example of a sales representative screen. FIG. 9 is a flowchart showing an example of the flow of operation processing according to a first modified example. FIG. 10 is a diagram showing an example of a shared display screen 600 according to a second modified example.

[0011] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0012] The description will be given in the following order: 1. Overview 2. Configuration 2-1. Configuration of server 20 2-2. Configuration of information processing terminal 100 3. Operational processing 4. Display screen example 5. Modifications 5-1. First modification 5-2. Second modification 6. Supplementary information

[0013] <1. Overview> As one embodiment of the present disclosure, a mechanism for prompting a customer to correct an answer will be described.

[0014] Fig. 1 is a configuration diagram of an information processing system 1 according to an embodiment of the present disclosure. As shown in Fig. 1, the information processing system 1 includes a customer terminal 10, a server 20, and a sales terminal 30. The customer terminal 10, the server 20, and the sales terminal 30 may be communicatively connected via a network 40.

[0015] In this embodiment, as an example, it is assumed that a customer (user) consults about insurance via the Internet. The customer uses a customer terminal 10 to communicate with a sales representative (another user). The sales representative uses a sales terminal 30 to communicate with the customer. Data communication between the customer terminal 10 and the sales terminal 30 can be performed via a server 20. The server 20 can perform control to appropriately display a customer screen on the customer terminal 10 and a sales representative screen on the sales terminal 30.

[0016] The server 20 may acquire the voice of the conversation between the customer and the sales representative, and collect customer information such as annual income, family composition, place of residence, and desired home purchase from the results of analyzing the voice. The server 20 may also assist the sales representative in presenting a life plan and insurance products. For example, the server 20 may display a basic life plan calculated from the collected customer information and information on recommended insurance products on the display screen of the sales terminal 30. It is expected that the sales representative will refer to the calculated life plan and, based on information obtained from the conversation with the customer and their own sales experience, present a more appropriate life plan and explain insurance products.

[0017] As mentioned above, it is desirable for customers to answer all questions honestly when interacting with sales representatives. However, it is possible that customers may give answers that are inaccurate due to embarrassment, vanity, or impulsive thoughts. It is a heavy burden for sales representatives to immediately determine the reliability of answers, and such judgment is difficult. However, if a life plan is presented without correcting the answers, the content will be of little use to the customer.

[0018] Therefore, the present disclosure makes it possible to prompt the user to correct their answer.

[0019] 2. Configuration Next, the configuration of each device included in the information processing system 1 will be described.

[0020] 2 is a block diagram showing an example of the configuration of the server 20 according to this embodiment. As shown in FIG. 2, the server 20 includes a communication unit 210, a control unit 220, and a storage unit 230.

[0021] (Communication Unit 210) The communication unit 210 has a transmission unit that transmits data to an external device and a reception unit that receives data from an external device. The communication unit 210 according to this embodiment may be communicatively connected to an external device or the Internet using, for example, a wired or wireless LAN (Local Area Network), Wi-Fi (registered trademark), Bluetooth (registered trademark), a mobile communication network (LTE (Long Term Evolution), 4G (fourth generation mobile communication system), 5G (fifth generation mobile communication system)), or the like.

[0022] For example, the communication unit 210 is connected to the customer terminal 10 and the sales terminal 30 via the network 40 for communication, and transmits and receives data.

[0023] (Control Unit 220) The control unit 220 functions as an arithmetic processing unit and a control device, and controls the overall operation of the server 20 in accordance with various programs. The control unit 220 is realized by an electronic circuit such as a CPU (Central Processing Unit) or a microprocessor. The control unit 220 may also include a ROM (Read Only Memory) that stores the programs to be used, arithmetic parameters, etc., and a RAM (Random Access Memory) that temporarily stores parameters that change as appropriate.

[0024] The control unit 220 can also function as an information acquisition unit 221 , a dialogue content analysis unit 222 , and a display control unit 223 .

[0025] The information acquisition unit 221 acquires a dialogue (spoken voice) between a customer and a sales representative that takes place between the customer terminal 10 and the sales terminal 30. The dialogue includes the customer's response to the sales representative's question, and the information acquisition unit 221 can acquire the customer's response information by analyzing the spoken voice. The customer's response information can be acquired not only from the customer's spoken voice but also from the sales representative's spoken voice. For example, if the customer answers "Yes" to the sales representative's question, "Is your annual income 5 million yen?", the information acquisition unit 221 acquires "annual income 5 million yen" as response information from the sales representative's spoken voice, "Is your annual income 5 million yen?" Note that the information acquisition unit 221 may also acquire survey information entered by the customer as text using the customer terminal 10 and acquire the customer's response information from the survey information.

[0026] The information acquisition unit 221 also acquires customer sensing data from the customer terminal 10. The customer sensing data is acquired from various sensors (camera, microphone, etc.) provided in the customer terminal 10. For example, the sensing data may include the customer's facial expression, line of sight, neck angle, body shaking, voice quality, voice tone, and pauses in speech. The information acquisition unit 221 outputs the customer's response information and the customer's sensing data when the customer responded to the dialogue content analysis unit 222.

[0027] The dialogue content analysis unit 222 analyzes the dialogue content between the customer and the sales representative. Specifically, the dialogue content analysis unit 222 performs voice analysis on the dialogue (spoken voice) between the customer and the sales representative acquired by the information acquisition unit 221, and acquires the customer's response (response information) as customer information. Note that the voice analysis may be performed by the information acquisition unit 221 or by another server. The response information is stored in the response information storage unit 233.

[0028] The dialogue content analysis unit 222 also has a function of calculating the reliability of response information based on the sensing data of the customer when the response information of the customer is acquired. The term "when the customer's response information is acquired" includes the time when the customer answers (utters). Furthermore, since the customer's response information may be acquired from the salesperson's speech (question), the term "when the customer's response information is acquired" may also include the time when the salesperson asks the question (utters). In other words, the customer's sensing data according to this embodiment may include sensing data before and after the customer's utterance. While the reliability calculation algorithm is not particularly limited, for example, the dialogue content analysis unit 222 calculates the reliability indicating the likelihood of the customer's response based on the tone of voice, facial expression, gaze, neck angle, body movement, etc. during the customer's response to determine whether the customer is providing an inaccurate answer. A trained AI (artificial intelligence) model may be used to calculate the reliability.

[0029] Furthermore, if attribute-specific answer information expected according to the customer's attributes (age, gender, educational background, occupation, place of residence, family structure, etc.) is prepared in advance, the dialogue content analysis unit 222 may also calculate the reliability by taking the attribute-specific answer information into consideration. For example, the greater the difference between the expected annual income corresponding to the customer's attributes and the annual income answered by the customer, the lower the calculated reliability will be. The attribute-specific answer information may be stored in, for example, the attribute information storage unit 232.

[0030] The dialogue content analysis unit 222 may also calculate the reliability by taking into account the content of the question asked by the salesperson. For example, when the question is about a place the customer wants to live in the future, a house, or future dreams, which are questions that the customer does not normally think deeply about, it is highly likely that the answer will be based on an impulse or that the customer himself does not understand. The dialogue content analysis unit 222 may take into account the content of the question and, for example, calculate a low reliability for answers to future predictions or vague questions.

[0031] The dialogue content analysis unit 222 stores the calculated reliability in the response information storage unit 233 in association with the response information.

[0032] The dialogue content analysis unit 222 also has a function of determining the importance of response information according to the dialogue content. In order to ease the customer's tension, the salesperson may engage in conversations unrelated to the main topic or digressions. Furthermore, the salesperson extracts certain information to be used from the information obtained from the customer to create a life plan, propose insurance products, etc., and does not necessarily use all of the information obtained. Therefore, it is desirable to obtain correct responses from the customer at least for the information to be used in creating a life plan, proposing insurance products, etc.

[0033] Here, it is assumed that the flow of an interview, such as an insurance consultation, has a certain pattern (which may be specific to each sales representative). For example, first, an icebreaker is conducted to ease the customer's tension. Next, conversation part A is conducted to obtain basic customer information, such as the customer's place of residence, family composition, marital status, occupation, age, health condition (medical history), annual income, savings balance, loan (current loan balance), and current lifestyle. Next, conversation part B is conducted to obtain future goals, such as the desired home, future dreams, ideal lifestyle, and children's educational environment. Next, conversation part C is conducted in which the sales representative creates a life plan (such as an income and expenditure simulation) based on the information obtained in the previous conversations, presents it to the customer, and explains it to them. Finally, conversation part D is conducted in which the sales representative gives advice, explains insurance products, etc. In this specification, a conversation part is a group of conversations for asking about or explaining specific information. Analysis of the conversation parts can be performed by the dialogue content analysis unit 222.

[0034] In the interview flow described above, the topic (subject) in conversation part C is the presentation of a life plan. To create such a life plan, at least some of the customer answers obtained in conversation part A, such as the customer's age, family structure, occupation, annual income, savings, loans, and lifestyle, are used. To optimize the life plan presented to the customer in conversation part C, these customer answers must be correct, and the dialogue content analysis unit 222 determines that answer information used in a conversation part, such as answer information used to create the life plan (information necessary for the topic in that conversation part), is important.

[0035] For example, the dialogue content analysis unit 222 can analyze the content of the dialogue between the customer and the sales representative in real time to understand what is currently being discussed. The dialogue content analysis unit 222 can determine the current conversation part and predict the next conversation part based on the content of the current conversation between the customer and the sales representative and pre-prepared information about the interview flow (specifically, the order of the conversation parts and the topics in each conversation part). The dialogue content analysis unit 222 determines the importance of answer information based on whether it is answer information used in the topic of the predicted next conversation part. Even if the answer information has a reliability lower than the above-mentioned threshold, if it is not information used in the topic (i.e., information with no importance), it will not hinder the conversation. However, if it is information that will be used (i.e., information with importance), it is necessary to obtain a correct answer from the customer for a useful conversation (presentation of a useful topic). Examples of information used in the topic (conversation) include information used to create a life plan. Furthermore, the dialogue content analysis unit 222 can detect a change in the conversation part (a change from the current conversation part to the next conversation part) in analyzing the content of the conversation between the customer and the sales representative.

[0036] In this embodiment, for answer information whose reliability is lower than the threshold and which is important, the customer is prompted to correct the answer, thereby enabling the customer to obtain a correct answer, thereby supporting the realization of a more effective interview. Furthermore, by prompting the customer to correct their answer before the next conversation part actually begins (for example, before creating a life plan), the sales representative can proceed with the interview without having to go back and forth, such as recreating a life plan.

[0037] Information about the interview flow is stored, for example, in the interview information storage unit 231. Information about the interview flow may be generated by the dialogue content analysis unit 222, or the sales representative may input an interview manual, talk script, etc. For example, the dialogue content analysis unit 222 learns the content of past conversations between sales representatives and customers, recognizes conversation parts, collects information such as the order of the conversation parts, topics in each conversation part, and the type of information used in each conversation part (topic), and structures the interview content.

[0038] The dialogue content analysis unit 222 may structure the interview content by learning the content of conversations between many sales representatives and customers, or may further structure the interview content personalized for each individual sales representative. When structuring the interview content, the dialogue content analysis unit 222 may also input industry knowledge, interview materials used by the sales representative, and the type of information to be input into a tool used to present topics (e.g., the type of information to be input into a life plan creation tool (program)). When learning the content of a conversation between a sales representative and a customer, the dialogue content analysis unit 222 may also learn the attributes of the customer. In other words, it is anticipated that the flow of the conversation may change depending on the customer's attributes, which increases the accuracy of structuring the interview content.

[0039] The dialogue content analysis unit 222 may learn the content of past dialogues between sales representatives and customers and generate attribute-specific answer information. For example, the dialogue content analysis unit 222 generates attribute-specific answer information by associating customer attributes with answer information that is likely to be given.

[0040] The dialogue content analysis unit 222 may calculate the influence of the answer information to be used on the result. Specifically, the dialogue content analysis unit 222 compares the results when answer information expected from the customer attributes (attribute-specific answer information) is used for the presented topic with the results when answer information answered by the customer is used, and determines that the greater the difference between the results, the greater the influence of the answer information on the result.

[0041] For example, in a conversation part in which a topic related to a life plan is presented to a customer, the dialogue content analysis unit 222 compares a life plan that uses an annual income estimated from customer attributes with a life plan that uses the annual income actually provided by the customer, and determines that a significant change in the plan content will have a large impact on the results. For example, a deviation of 1 million yen in annual income can significantly change the plan content, such as a deviation of 5 million yen in the home purchase budget. Furthermore, in a conversation part in which a topic related to insurance products is presented to a customer, the greater the difference in the insurance product recommended to the customer, the greater the impact on the results. The dialogue content analysis unit 222 may learn the content of past conversations between sales representatives and customers and generate a learning model that calculates the magnitude of the impact. Even for information with a high impact, if the reliability is low, it is desirable to have the customer correct the answer to the correct answer.

[0042] Existing technologies such as LLM (Large Scale Language Model) may be used for structuring the interview content, generating attribute-specific response information, and calculating the influence as described above.

[0043] The display control unit 223 controls the display of a customer screen on the customer terminal 10 and the display of a sales representative screen on the sales terminal 30. Specifically, the display control unit 223 transmits display information of the customer screen to the customer terminal 10, and also transmits display information of the sales representative screen to the sales terminal 30.

[0044] Furthermore, the display control unit 223 controls the display on the customer terminal 10 of a message prompting the customer to correct answer information whose reliability is lower than a threshold and which is important (specifically, used in topics) according to the analysis result by the dialogue content analysis unit 222. When an answer is corrected, the sales representative is notified of this and can use the correct answer.

[0045] (Storage Unit 230) The storage unit 230 is realized by a ROM that stores programs and calculation parameters used in the processing of the control unit 220, and a RAM that temporarily stores parameters that change as needed.

[0046] The memory unit 230 stores an interview information memory unit 231 that stores information about interviews, such as dialogue audio from past interviews and information about the flow of the interviews (structured information about the interview content), an attribute information memory unit 232 that stores information that can be associated with customer attributes (e.g., attribute-specific answer information), and a response information memory unit 233 that stores customer answer information.

[0047] <<2-2. Configuration of Information Processing Terminal 100>> The customer terminal 10 and the sales terminal 30 can be realized by the information processing terminal 100. The information processing terminal 100 may be, for example, a smartphone, a tablet terminal, a PC (personal computer), or an HMD (head mounted display). The basic configuration of the information processing terminal 100 applied to the customer terminal 10 and the sales terminal 30 will be described below.

[0048] 3 is a block diagram showing an example of a configuration of an information processing terminal 100 according to an embodiment of the present disclosure. As shown in FIG. 3, the information processing terminal 100 includes a communication unit 110, an operation display unit 120, an audio input unit 130, a control unit 140, a storage unit 150, an audio output unit 160, and a camera 170.

[0049] (Communication Unit 110) The communication unit 110 has a transmission unit that transmits data to an external device and a reception unit that receives data from the external device. The communication unit 110 according to this embodiment may be communicatively connected to an external device or the Internet using, for example, a wired or wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), a mobile communication network (LTE, 4G, 5G), or the like.

[0050] (Operation display unit 120) The operation display unit 120 functions as an operation unit that accepts operation input by the user and outputs the input information to the control unit 140, and as a display unit that displays various screens. The operation display unit 120 may be realized by, for example, a touch panel display. Note that the operation unit and the display unit may be configured as separate entities.

[0051] (Audio Input Unit 130) The audio input unit 130 collects the user's speech and outputs the audio data to the control unit 140. The audio input unit 130 can be realized by, for example, a microphone.

[0052] (Audio Output Unit 160) The audio output unit 160 has a function of outputting audio. The audio output unit 160 can be realized by, for example, a speaker or earphones.

[0053] (Camera 170) The camera 170 captures an image of a subject and outputs the captured image to the control unit 140. For example, the camera 170 captures an image of a customer who is having a conversation with a sales representative while looking at a customer screen displayed on the operation display unit 120 of the customer terminal 10, and outputs the captured image to the control unit 140.

[0054] (Control Unit 140) The control unit 140 functions as an arithmetic processing unit and a control device, and controls the overall operation of the information processing terminal 100 in accordance with various programs. The control unit 140 is realized by an electronic circuit such as a CPU or a microprocessor. The control unit 140 may also include a ROM for storing the programs to be used, arithmetic parameters, etc., and a RAM for temporarily storing parameters that change as appropriate.

[0055] For example, the control unit 140 controls the communication unit 110 to transmit data of the speech of the customer or the sales representative input from the speech input unit 130 to the server 20. The control unit 140 also controls the speech of the customer or the sales representative received from the communication unit 110 to be output from the speech output unit 160. The control unit 140 also controls the operation display unit 120 to display a customer screen or a sales representative screen using information received from the server 20. The control unit 140 also controls the communication unit 110 to transmit information input using the operation display unit 120 to the server 20. The control unit 140 also controls the communication unit 110 to transmit an image of the customer captured by the camera 170 to the server 20. The captured image of the customer is an example of customer sensing data. The customer's speech is also analyzed in the server 20 and used as customer sensing data (e.g., tone of voice, pauses in speech, etc.).

[0056] (Storage Unit 150) The storage unit 150 is realized by a ROM that stores programs and calculation parameters used in the processing of the control unit 140, and a RAM that temporarily stores parameters that change as needed.

[0057] The basic configuration of the information processing terminal 100 has been specifically described above. Note that the configuration of the information processing terminal 100 according to the present disclosure is not limited to the example shown in Fig. 3. For example, the information processing terminal 100 does not necessarily have all of the configurations shown in Fig. 3. Furthermore, the information processing terminal 100 may be realized by a plurality of devices.

[0058] 3. Operation Processing FIG. 4 is a flowchart showing an example of the flow of operation processing according to this embodiment.

[0059] 4, first, the server 20 performs interview pre-processing by the control unit 220 (step S103). Specifically, the control unit 220 learns dialogue voices from past interviews and structures the interview content.

[0060] Next, the server 20 acquires, via the information acquisition unit 221, the dialogue voice of the interview between the sales representative and the customer that is being conducted in real time, as well as sensing data of the customer (step S106). The dialogue voice also includes the customer's responses to the sales representative's questions. The information acquisition unit 221 may perform speech recognition on the dialogue voice to convert the dialogue voice into text, and may further perform natural language processing to acquire the customer's response information. The sensing data of the customer includes the customer's facial expression, line of sight, body movements, neck angle, tone of voice, pauses in speech, and the like, when the sales representative is asking a question or when the customer is answering the sales representative's question. The dialogue voice and sensing data may be acquired continuously.

[0061] Next, the server 20 calculates the reliability of the corresponding answer information based on the sensing data when the customer is answering (step S109) using the dialogue content analysis unit 222. The dialogue content analysis unit 222 calculates the reliability of the answer information acquired from the dialogue voice being carried out in real time by referring to the sensing data.

[0062] Next, the server 20 determines, via the dialogue content analysis unit 222, whether a conversation part will switch (step S112). Specifically, the dialogue content analysis unit 222 determines (detects) whether a conversation part will switch based on information about the conversation flow (such as the order of the conversation parts and the topics in each conversation part) indicated in the structured information about the interview content prepared in advance and the content of the currently ongoing dialogue. For example, the dialogue content analysis unit 222 may determine (detect) a conversation part switch when all planned questions in the current conversation part have been asked or based on the words of the sales representative. The dialogue content analysis unit 222 continuously analyzes the dialogue content in real time. If a conversation part has not switched (step S112 / No), the dialogue content analysis unit 222 repeats the processes of steps S106 to S112.

[0063] Next, if it is determined that the conversation part is switching (step S112 / Yes), the dialogue content analysis unit 222 checks whether the reliability of the answer information to be used in the next topic is lower than a threshold (step S115). The type of answer information to be used (presented) in the next conversation part that begins after the conversation part has switched is obtained from the structured information of the interview content generated in the interview pre-processing. The dialogue content analysis unit 222 checks whether the reliability of the answer information to be used in the next topic (i.e., important) is lower than a threshold. If there are multiple pieces of answer information to be used, the dialogue content analysis unit 222 checks whether the reliability of each piece of answer information is lower than a threshold.

[0064] Next, if the reliability of the answer information to be used is lower than the threshold (step S115 / Yes), the dialogue content analysis unit 222 determines whether the answer information will have a large impact on the results when used (step S118).

[0065] Next, if it is determined that the impact is large (Yes in step S118), the server 20 controls the display to prompt the customer to correct the answer information by the display control unit 223 (step S121). This is expected to encourage the customer to inform the sales representative of the need to correct the answer.

[0066] On the other hand, if the reliability of the answer information used is not lower than the threshold value (step S115 / No), or if the reliability of the answer information used is lower than the threshold value but the impact is not large (step S118 / No), the server 20 does not prompt the customer to correct the answer information.

[0067] In this way, when the reliability of answer information (important answer information) to be used in the next topic is low and has a large impact on the results, the server 20 prompts the customer to correct the answer information, thereby obtaining a correct answer and realizing a useful interview using the correct answer. Even if a customer gives an answer that is incorrect due to embarrassment or other reasons, the system's prompt to correct the answer is expected to prompt the customer to correct the answer to the correct one. Furthermore, the sales representative does not need to determine the reliability of the answer, thereby reducing the burden on the sales representative. Furthermore, when prompting the customer to correct the answer, the display control unit 223 adds an explanation that the answer will be used in the next topic and that if the answer is incorrect, the answer will have a large impact on the results, thereby motivating the customer to give a correct answer. On the other hand, by not prompting the customer to correct the answer for answers that have a small impact on the results, i.e., minor lies (e.g., slightly understating one's annual income), the burden on the customer can be reduced.

[0068] The server 20 then repeats the processes of steps S106 to S124 until the interview is completed (step S124 / No). If the interview is completed (step S124 / Yes), this operation process is completed.

[0069] The above describes the operation processing according to this embodiment. Note that the operation processing shown in Fig. 4 is an example, and this embodiment is not limited to this. The operation processing shown in Fig. 4 does not have to be performed in the order shown in Fig. 4. For example, it is possible to skip some processes, reverse the order of some processes, or perform some processes in parallel.

[0070] Specifically, for example, the process shown in step S118 is not essential and may be skipped. In this system, by prompting the customer to correct the answer information when the reliability of at least the answer information used in the topic is low, rather than all the answer information with low reliability, it is possible to realize a useful interview while also reducing the burden on the customer.

[0071] Furthermore, the server 20 may prompt the customer to revise their answer if the relationship of trust between the customer and the sales representative is sufficient. This is because if the relationship of trust between the customer and the sales representative is not sufficient, the customer may not revise their answer even when prompted. The control unit 220 of the server 20 may determine whether the relationship of trust between the customer and the sales representative is sufficient, for example, based on the customer's facial expression, posture, gestures, tone of voice, etc. while talking with the sales representative. The control unit 220 may also determine whether the relationship of trust between the customer and the sales representative is sufficient by taking into account the number of meetings between the customer and the sales representative. The timing of determining whether the relationship of trust between the customer and the sales representative is sufficient may be before step S115, between step S115 and step S118, or after step S118.

[0072] Furthermore, the timing for calculating the reliability in step S109 may be after step S112. This is because it is possible that a relationship of trust is built as the customer continues to talk with the sales representative and the customer changes his or her answer on his or her own, or that the customer initially does not have a clear idea of ​​the conditions for the house he or she would like to live in in the future and answers on impulse, but as the interview progresses, specific wishes emerge.

[0073] 4. Display Screen Examples Next, examples of display screens for a customer screen and a sales representative screen when prompting a customer to correct response information will be described with reference to the drawings.

[0074] (1) Figure 5 shows an example of a customer screen and a sales representative screen according to this embodiment. The customer side display screen 400 shown in the upper part of Figure 5 is displayed on the customer terminal 10. The sales representative side display screen 500 shown in the lower part of Figure 5 is displayed on the sales terminal 30.

[0075] The customer-side display screen 400 may be displayed, for example, when the conversation part switches and the reliability of the answer information to be used in the next topic is low. The customer-side display screen 400 displays the answer information to be used in the next topic, "annual income of 4.5 million yen," and further displays a message urging correction, saying, "If there is a discrepancy, please inform the sales representative." By also being shown what the answer information will be used for, the customer, wanting to have the correct life plan made, can correct the answer by saying, for example, "Actually, my annual income is a little lower, at 4 million yen."

[0076] The server 20 may prompt the customer to revise their answer while notifying the sales representative accordingly. The sales representative's display screen 500 presents the customer answer to be used in the next topic and its reliability. Furthermore, the item among the customer answers for which the customer is being prompted to revise their answer is displayed as "Customer Suggested." Here, the threshold for determining low reliability is set to 50%, and the customer is prompted to revise the customer answer "Annual Income: 4.5 million yen" with a reliability of less than 50%. When the "Customer Suggested" message is displayed, the sales representative can create an atmosphere in which the customer can easily revise their answer by slowing down their speaking speed, making a side conversation, or taking a breath before moving on to the next topic.

[0077] (2) Figure 6 shows other examples of the customer screen and the sales representative screen according to this embodiment. The customer display screen 410 shown in the upper part of Figure 6 is displayed on the customer terminal 10. The sales representative display screen 500 shown in the lower part of Figure 6 is the same as the example screen shown in the lower part of Figure 5 and is displayed on the sales terminal 30.

[0078] In addition to the example screen shown in the upper part of Fig. 5, the customer-side display screen 410 also includes information about the magnitude of the impact, such as "For example, a discrepancy of 500,000 yen will have a significant impact on your life plan." This further increases the customer's motivation to revise their answers.

[0079] (3) Figure 7 shows other examples of the customer screen and the sales representative screen according to this embodiment. The customer side display screen 420 shown in the upper part of Figure 7 is displayed on the customer terminal 10. The sales representative side display screen 520 shown in the lower part of Figure 7 is displayed on the sales terminal 30.

[0080] The customer-side display screen 420 differs from the screen examples shown in the upper part of FIG. 5 or the upper part of FIG. 6 in that it includes additional text prompting the customer to tap on the item they wish to modify. In the example shown in FIG. 7, the "annual income of 4.5 million yen" section of the customer-side display screen 420 is displayed as a button that can be tapped. Some customers may find it difficult to verbally express their desire to modify their answer. In this embodiment, a tap operation allows the customer to communicate their intention to modify their answer to the sales representative. Note that while only one item is presented here as an example, if there are other items with reliability levels lower than a threshold (e.g., 50%), multiple items may be presented.

[0081] When the customer taps an item on the sales representative side display screen 520, a message is displayed informing the customer of their intention to revise their answer, such as "It appears the customer wishes to revise their answer regarding 'annual income.'" The sales representative can make it easier for the customer to discuss revising their answer by asking, for example, "Would you like to revise your annual income?"

[0082] (4) Fig. 8 is a diagram showing another example of a sales representative screen. A sales representative side display screen 530 shown in Fig. 8 is displayed on the sales terminal 30. On the customer side, a customer side display screen shown in any of Figs. 5 to 7 is displayed.

[0083] The control unit 220 of the server 20 may estimate whether the customer intends to make corrections based on sensing data obtained by sensing the customer's reaction and may notify the sales representative of the estimated result. For example, the control unit 220 estimates the intention to make corrections based on sensing data such as the customer's line of sight, facial expression, body movement, and head tilt when the customer-side display screen shown in any of Figures 5 to 7 is displayed.

[0084] On the sales representative's display screen 530, a notification such as "It appears that the customer has revised their answer regarding 'annual income'" is displayed, and the sales representative is informed of the estimated result of the customer's intention to revise their answer. The sales representative can then make it easier for the customer to discuss the revision by saying, for example, "We can revise the annual income we asked you about earlier. Would you like to do that?"

[0085] If multiple items are displayed on the customer display screen, the control unit 220 may infer which item the customer is having trouble with from the way the customer looks at the screen. Even if the control unit 220 does not know which item the customer is having trouble with, it may notify the sales representative, for example, "It looks like there is an item you would like to correct," and have the sales representative find out which item the customer wants to correct.

[0086] 5. Modifications Next, modifications of this embodiment will be described.

[0087] <<5-1. First Modification>> In the embodiment described above, interview pre-processing is performed and structured information of the interview content is generated as explained with reference to Figure 4, but this embodiment is not limited to this. In the first modification, a case where interview pre-processing is not performed will be described.

[0088] 9 is a flowchart showing an example of the flow of the operation process according to the first modified example. As shown in Fig. 9, first, the server 20 acquires, via the information acquisition unit 221, conversational voices during a real-time interview between a sales representative and a customer, and sensing data of the customer (step S203).

[0089] Next, the server 20 calculates the reliability of the corresponding answer information based on the sensing data when the customer is answering using the dialogue content analysis unit 222 (step S206).

[0090] Next, the dialogue content analysis unit 222 checks whether the reliability of the response information used in the current topic is lower than a threshold (step S209). If structured information about the interview content is not prepared in advance, the content and order of the conversation parts are unknown, making it impossible to recognize when the conversation parts change. In this modified example, the sales representative's speech, such as "If your annual income is 4.5 million yen, what is your life plan...?", is analyzed in real time and compared with the customer's response information obtained from the dialogue up to that point. The dialogue content analysis unit 222 then checks whether the reliability of the customer's response information regarding, for example, "annual income," is lower than a threshold.

[0091] Next, if the reliability of the answer information to be used is lower than the threshold (step S209 / Yes), the server 20 controls the display to prompt the customer to correct the answer information by the display control unit 223 (step S212). This is expected to encourage the customer to inform the sales representative of the correction of the answer.

[0092] The server 20 then repeats the processes of steps S203 to S215 until the interview is completed (step S215 / No). If the interview is completed (step S215 / Yes), this operation process is completed.

[0093] As described above, if the structured information of the interview content is not updated, it is impossible to grasp the transition of conversation parts or the answer information to be used in the next topic. However, by analyzing the dialogue voice in real time, it is possible to determine the answer information to be used in the current topic (important answer information). If the reliability of the answer information is low, the customer can be prompted to correct it, thereby realizing a useful interview. The customer screen displayed on the customer terminal 10 may display a message such as, "We are explaining your life plan using the information you provided earlier, 'annual income of 4.5 million yen.' If there is a discrepancy, please inform a sales representative." This prompts the customer to correct the answer information. The server 20 can display the prompt to correct the customer's answer information simultaneously or approximately simultaneously with the acquisition of the dialogue content. This is expected to enable the correct answer to be obtained as quickly as possible, resulting in a more effective interview.

[0094] <<5-2. Second Modification>> In the above-described embodiment, it is assumed that the customer and the sales representative do not meet face-to-face in real life, but rather converse over a network using each of the information processing terminals 100. On the other hand, it is also conceivable that the present system can be applied to a real-life face-to-face interview between a customer and a sales representative.

[0095] In this case, it is expected that the sales representative will share with the customer the screen displayed on the display unit of the sales terminal 30 or the store's counter terminal (not shown), and display the life plan, etc. on the screen and explain it to the customer.

[0096] Considering that the sales representative and the customer share the display screen, this modification does not explicitly guide the customer to make corrections. Instead, for example, only answer information used in the topic that has low reliability may be presented on the display screen. FIG. 10 illustrates an example of a shared display screen 600 according to the second modification. Here, it is assumed that the server 20 performs interview preprocessing and prepares structured information about the interview content in advance. When the server 20 detects a change in conversation part, it presents only answer information to be used in the next topic whose reliability is lower than a threshold. Specifically, as shown in the notification 602 pop-up displayed on the shared display screen 600 in FIG. 10 , the notification 602 does not include any text encouraging the customer to make corrections, but instead presents answer information to be used in the next topic. Because the sales representative understands that answer information with low reliability is presented as in the notification 602, they can create an atmosphere that encourages corrections by saying something like, "May I proceed without correcting the annual salary?", which is expected to lead to a correct answer.

[0097] The server 20 may also display a list of answer information to be used in the next topic on the shared display screen, and display answer information with a reliability lower than a threshold in a display format that only the sales representative can understand (for example, by marking it, etc.). This provides an atmosphere in which the sales representative can easily make corrections, and is expected to obtain the correct answer.

[0098] On the other hand, if interview pre-processing is not performed and structured information about the interview content is not prepared in advance, the server 20 may analyze the dialogue voice in real time, and if the reliability of the obtained answer information is lower than a threshold, display answer information with low reliability such as "Customer's answer: ..." on the shared display screen. Furthermore, the server 20 may analyze the dialogue voice in real time, identify answer information used in the current topic, and if the reliability of the answer information is lower than a threshold, display a notification such as "We are making a proposal using the information you just answered ..." on the shared display screen. The salesperson understands that this display indicates a case where the reliability of the answer information is low, and providing an atmosphere that makes it easy to make corrections can be expected to lead to a correct answer.

[0099] <6. Supplementary Information> Although preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, the present technology is not limited to such examples. It is clear that a person skilled in the art of the present disclosure can conceive of various modified or altered examples within the scope of the technical idea described in the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

[0100] For example, one or more computer programs can be created to cause hardware such as a CPU, ROM, and RAM built into the above-described server 20 or information processing terminal 100 to perform the functions of the server 20 or the information processing terminal 100. Also provided is a computer-readable storage medium storing the one or more computer programs.

[0101] In the above-described embodiment, an example has been described in which the customer's speech is analyzed to acquire the customer's response information, but the present invention is not limited to this, and the customer's response information may be acquired from information entered as text by the customer. In this case, the information acquisition unit 221 may acquire sensing data while the customer is entering text.

[0102] In addition, when the server 20 obtains response information from the questionnaire information entered by the customer, the reliability of the response information can be calculated based on the customer's facial expression and gaze when entering the questionnaire, as well as the time it took to enter the response and whether or not the response was corrected.

[0103] The functional configuration of the server 20 may also be provided in the information processing terminal 100. Specifically, for example, the functional configuration of the server 20 may be realized by an application that can run on each of the customer terminal 10 and the sales terminal 30.

[0104] Furthermore, the effects described herein are merely descriptive or exemplary and are not limiting. In other words, the technology according to the present disclosure may achieve other effects that will be apparent to those skilled in the art from the description of this specification, in addition to or in place of the above-described effects.

[0105] Note that the present technology can also be configured as follows. (1) An information processing method including: a processor acquiring answer information of a user; acquiring sensing data of the user; calculating the reliability of the answer information based on the sensing data of the user when the answer information was acquired; determining the importance of the answer information based on the content of a dialogue between the user and another user; and controlling display to prompt the user to correct the answer information according to the importance and reliability of the answer information. (2) The information processing method according to (1), in which the processor controls display of answer information that is important and has a reliability lower than a threshold value to prompt the user to correct the answer information. (3) The information processing method according to (1) or (2), in which the processor analyzes the content of the dialogue in real time to determine the importance of the answer information. (4) The information processing method according to (1), wherein, in analyzing the content of the dialogue, the processor detects a change in conversation part by referring to structured information of the interview content generated by learning past interview content, and upon detecting a change in the conversation part, identifies answer information to be used in the next conversation part from the structured information of the interview content and determines that the answer information to be used in the next conversation part is important. (5) The information processing method according to (4), wherein, when the reliability of the answer information determined to be important is lower than a threshold, the processor controls to display a message on a display screen of the information processing terminal used by the user prompting the user to correct the answer information. (6) The information processing method according to (5), wherein the message prompting the user to correct the answer information includes a message that the answer information will be used in the next topic. (7) The information processing method according to any one of (4) to (6), wherein the processor further calculates the magnitude of the impact that the answer information will have on the results if it is used, and performs display control to prompt the user to correct the answer information that is determined to be important, has a reliability lower than a threshold, and has a large impact.(8) The information processing method according to (1), wherein the processor, in analyzing the content of the dialogue, grasps answer information used in a conversation part, determines that the answer information used in the conversation part is important, and if the reliability of the answer information determined to be important is lower than a threshold, controls a display on a display screen of an information processing terminal used by the user to prompt the user to correct the answer information simultaneously or approximately simultaneously with acquisition of the content of the dialogue. (9) The information processing method according to any one of (1) to (8), wherein, while prompting the user to correct the answer information, the processor controls a display screen of an information processing terminal used by the other user to display a message indicating that the answer information is being presented to the user. (10) The information processing method according to any one of (1) to (9), wherein the user's answer information is obtained from dialogue voice between the user and the other user. (11) A program that causes a computer to function as a control unit that performs the following processes: acquiring answer information of a user, acquiring sensing data of the user, calculating the reliability of the answer information based on the sensing data of the user when the answer information was acquired, determining the importance of the answer information based on the content of a dialogue between the user and another user, and controlling display to prompt the user to correct the answer information according to the importance and reliability of the answer information. (12) An information processing device comprising a control unit that performs the following processes: acquiring answer information of a user, acquiring sensing data of the user, calculating the reliability of the answer information based on the sensing data of the user when the answer information was acquired, determining the importance of the answer information based on the content of a dialogue between the user and another user, and controlling display to prompt the user to correct the answer information according to the importance and reliability of the answer information.

[0106] REFERENCE SIGNS LIST 1 Information processing system 10 Customer terminal 20 Server 210 Communication unit 220 Control unit 221 Information acquisition unit 222 Dialogue content analysis unit 223 Display control unit 230 Storage unit 231 Interview information storage unit 232 Attribute information storage unit 233 Response information storage unit 30 Sales terminal 40 Network

Claims

1. An information processing method including: a processor acquiring answer information of a user; acquiring sensing data of the user; calculating the reliability of the answer information based on the sensing data of the user at the time of acquiring the answer information; determining the importance of the answer information based on the content of a dialogue between the user and another user; and controlling display to prompt the user to correct the answer information according to the importance and reliability of the answer information.

2. The information processing method according to claim 1, wherein the processor controls display of answer information that is important and has a reliability lower than a threshold value to prompt the user to correct the answer information.

3. The information processing method according to claim 1, wherein the processor analyzes the dialogue content in real time to determine the importance of the response information.

4. The information processing method of claim 1, wherein the processor, in analyzing the dialogue content, detects a change in conversation part by referring to structured information of the interview content generated by learning past interview content, and when detecting a change in conversation part, identifies answer information to be used in the next conversation part from the structured information of the interview content and determines that the answer information to be used in the next conversation part is important.

5. An information processing method as described in claim 4, wherein the processor controls the display screen of the information processing terminal used by the user to display a message prompting the user to correct the answer information if the reliability of the answer information determined to be important is lower than a threshold value.

6. The information processing method according to claim 5, wherein the display prompting the user to correct the answer information includes a message that the answer information will be used in the next topic.

7. The information processing method of claim 4, wherein the processor further calculates the magnitude of the impact that the answer information will have on the result if it is used, and performs display control to prompt the user to correct answer information that is determined to be important, has a reliability lower than a threshold, and has a large impact.

8. The information processing method according to claim 1, wherein the processor, in analyzing the content of the dialogue, grasps answer information used in the conversation part, determines that the answer information used in the conversation part is important, and if the reliability of the answer information determined to be important is lower than a threshold, controls the display screen of the information processing terminal used by the user to display a message prompting the user to correct the answer information simultaneously or approximately simultaneously with the acquisition of the content of the dialogue.

9. An information processing method as described in claim 1, wherein the processor, while prompting the user to correct the answer information, controls the display screen of the information processing terminal used by the other user to display a message indicating that the answer information is being presented to the user.

10. The information processing method according to claim 1, wherein the user's response information is obtained from a conversation between the user and another user.

11. A program that causes a computer to function as a control unit that performs the following processes: acquiring user response information; acquiring sensing data of the user; calculating the reliability of the response information based on the sensing data of the user when the response information was acquired; determining the importance of the response information based on the content of a dialogue between the user and another user; and controlling the display to prompt the user to correct the response information according to the importance and reliability of the response information.

12. An information processing device comprising a control unit that performs the following processes: a process of acquiring user answer information; a process of acquiring sensing data of the user; a process of calculating the reliability of the answer information based on the sensing data of the user when the answer information was acquired; a process of determining the importance of the answer information based on the content of a dialogue between the user and another user; and a process of controlling display to prompt the user to correct the answer information according to the importance and reliability of the answer information.

Images