Information processing apparatus and information processing method

Abstract

Enhance the persuasive power when proposing a product. 【Solution means】The information processing apparatus 1 includes a reception unit 133 that receives an inquiry regarding a product from the user terminal 2, a first acquisition unit 134 that refers to a product DB 121 storing product information and acquires product information of a product corresponding to the inquiry, a plurality of products, and a customer service method DB 122 that stores customer service method information regarding a customer service method corresponding to each of the plurality of products, in which each of the plurality of products is associated with a product related to each of the plurality of products. The apparatus refers to the customer service method DB 122, specifies customer service method information corresponding to the product indicated by the acquired product information, and based on the specified customer service method information, specifies an associated product related to the product indicated by the product information by a specifying unit 135. The apparatus refers to the product DB 121, acquires product information corresponding to the associated product by a second acquisition unit 136, and has an output unit 138 that outputs the product information acquired by the first acquisition unit 134 and the product information acquired by the second acquisition unit 136 to the user terminal 2.

Description

Technical Field

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

Background Art

[0002] In recent years, systems for chatting with users using large language models (LLMs) have been provided. For example, Patent Document 1 discloses a system that conducts conversations with users and proposes products to the users.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional method of proposing products using large language models, in response to a product inquiry from a user, it merely stops at introducing the product. For this reason, like a store clerk who is good at selling products, in addition to the specifications such as the size of the product in response to an inquiry from the user, explanations such as how to use the product that can lead to the user imagining the usage image of the product and increasing the purchasing desire are added for proposal, or products other than the product in response to an inquiry from the user are also proposed, and it is required to enhance the persuasive power when proposing products.

[0005] Therefore, the present invention has been made in view of these points, and an object thereof is to enhance the persuasive power when proposing products.

Means for Solving the Problems

[0006] A first aspect of the present invention is an information processing apparatus. This information processing apparatus includes a reception unit that receives an inquiry regarding a product from a user terminal used by a user, a first acquisition unit that refers to a product database storing product information regarding each of a plurality of products and acquires the product information of the product corresponding to the inquiry, a plurality of products, and a customer service method information database storing customer service method information regarding the customer service method corresponding to each of the plurality of products, which associates each of the plurality of products with related products related to each of the plurality of products. The apparatus refers to the customer service method information database, identifies the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit, and based on the identified customer service method information, a specifying unit that specifies related products related to the product indicated by the product information acquired by the first acquisition unit, a second acquisition unit that refers to the product database and acquires the product information corresponding to the related products specified by the specifying unit, and an output unit that outputs the product information acquired by the first acquisition unit and the product information acquired by the second acquisition unit to the user terminal.

[0007] The first acquisition unit may acquire the product information of the product corresponding to the inquiry by inputting the inquiry to a large language model capable of referring to the product database and the customer service method information database and acquiring an answer including the product information corresponding to the inquiry output from the large language model. The specifying unit may identify the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit by inputting instruction information instructing the acquisition of the customer service method information corresponding to the product information acquired by the first acquisition unit to the large language model and acquiring an answer including the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit from the large language model. The second acquisition unit may acquire the product information corresponding to the related products by inputting instruction information instructing the acquisition of the product information corresponding to the related products specified by the specifying unit to the large language model and acquiring an answer including the product information output from the large language model.

[0008] The information processing device may have a generation unit that creates an answer sentence corresponding to the inquiry, which includes the product information acquired by the first acquisition unit and the product information acquired by the second acquisition unit, and instructs the large language model to create an answer sentence that promotes the purchase of the product indicated by the product information acquired by the first acquisition unit and the related product indicated by the product information acquired by the second acquisition unit, and acquires the answer sentence from the large language model, and the output unit may output the answer sentence generated by the generation unit.

[0009] The customer service method information may be associated with salesperson identification information for identifying the salesperson who provided customer service corresponding to the customer service method, and the specifying unit may specify the salesperson identification information associated with the specified customer service method information, and may associate the specified salesperson identification information with a usage record indicating that the customer service method information has been used.

[0010] The customer service method database may store the customer service method information in association with attribute information indicating the attributes of the user corresponding to the customer service method corresponding to the customer service method information, and the specifying unit may specify the attributes of the user using the user terminal, and with reference to the customer service method database, may specify the customer service method information corresponding to the attributes of the specified user and the product indicated by the product information acquired by the first acquisition unit.

[0011] The specifying unit may specify the attributes of the user by causing a large language model to estimate the attributes of the user based on the conversation history between the large language model and the user using the user terminal.

[0012] The information processing device may include an extraction unit that acquires text information indicating the speech of a salesperson in a store and extracts text information corresponding to a customer service method from the acquired text information, and a storage control unit that stores the text information extracted by the extraction unit in the customer service method database as the customer service method information.

[0013] The memory control unit identifies the category of products or services handled by the store clerk, and also identifies the category of products or services corresponding to the extracted text information. The memory control unit may store the extracted text information in the customer service method database as the customer service method information on the condition that the category of products or services handled by the store clerk matches the category of products or services corresponding to the extracted text information.

[0014] The extraction unit may extract information for identifying a product from the acquired text information. The memory control unit may associate the information for identifying the product extracted by the extraction unit with the text information corresponding to the customer service method and store the result in the customer service method database as the customer service method information.

[0015] The memory control unit refers to user information in which user identification information for identifying each of a plurality of users is associated with attribute information indicating the attributes of the users, identifies the attribute information associated with the user identification information of the user with whom the store clerk interacted in the store, and may associate the customer service method information with the identified attribute information and store the result in the customer service method database.

[0016] A second aspect of the present invention is an information processing method. This information processing method includes steps executed by a computer: receiving an inquiry about a product from a user terminal used by a user; referring to a product database storing product information for each of a plurality of products, and obtaining the product information of the product corresponding to the inquiry; referring to a customer service method database storing customer service method information regarding customer service methods corresponding to each of the plurality of products, which associates the plurality of products with related products respectively related to each of the plurality of products, identifying the customer service method information corresponding to the product indicated by the obtained product information, and based on the identified customer service method information, identifying related products related to the product indicated by the obtained product information; referring to the product database and obtaining the product information corresponding to the identified related products; and outputting the product information of the product corresponding to the obtained inquiry and the product information of the obtained related products to the user terminal.

Advantages of the Invention

[0017] According to the present invention, there is an effect that the persuasive power at the time of proposing a product can be enhanced.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0019] [Overview of Information Processing Apparatus 1] FIG. 1 is a diagram showing an overview of the information processing apparatus 1. The information processing apparatus 1 is a computer for serving users who use a store in a store or the like that sells products. The information processing apparatus 1 is communicably connected to a user terminal 2 used by the user.

[0020] First, the information processing apparatus 1 receives an inquiry about a product from the user ((1) in FIG. 1). The information processing apparatus 1 refers to a product DB (database) that stores product information about each of a plurality of products, and acquires the product information of the product corresponding to the inquiry ((2) in FIG. 1).

[0021] The information processing apparatus 1 refers to a customer service method DB that stores customer service method information regarding a customer service method corresponding to a product, which associates a product with related products related to the product, and specifies the customer service method information corresponding to the product indicated by the acquired product information ((3) in FIG. 1). The information processing apparatus 1 specifies related products related to the product corresponding to the inquiry based on the specified customer service method information ((4) in FIG. 1).

[0022] The information processing apparatus 1 refers to the product DB and acquires the product information of the specified related products ((5) in FIG. 1). The information processing apparatus 1 outputs the product information of the product and the related products to the user terminal 2 ((6) in FIG. 1). By doing so, the information processing apparatus 1 can propose not only the specifications such as the size of the product for which an inquiry has been received from the user, but also explanations such as how to use the product that can lead the user to imagine the usage image of the product and enhance the purchasing desire, and can also propose products other than the product for which an inquiry has been received from the user. Therefore, the information processing apparatus 1 can enhance the appeal power when proposing products.

[0023] [Functional Configuration of Information Processing Apparatus 1] Subsequently, the functions of the information processing apparatus 1 will be described. FIG. 2 is a diagram showing the functional configuration of the information processing apparatus 1. The information processing apparatus 1 includes a communication unit 11, a storage unit 12, and a control unit 13.

[0024] The communication unit 11 is a communication interface for transmitting and receiving data to and from external devices such as the user terminal 2 via a communication network such as the Internet or a mobile phone line. The storage unit 12 is a storage medium for storing various types of data, and includes a ROM (Read Only Memory), a RAM (Random Access Memory), and a hard disk, etc. The storage unit 12 stores the programs executed by the control unit 13. The storage unit 12 stores a program that causes the control unit 13 to function as an extraction unit 131, a storage control unit 132, a reception unit 133, a first acquisition unit 134, a specification unit 135, a second acquisition unit 136, a generation unit 137, and an output unit 138.

[0025] In addition, the storage unit 12 stores a product DB 121 and a customer service method DB 122. The product DB 121 is a database that stores product information regarding each of a plurality of products, and is set to be referable by a large language model described later. Specifically, the product DB 121 stores by associating product identification information for identifying a product with product information. FIG. 3 is a diagram showing an example of the product DB 121. As shown in FIG. 3, the product DB 121 stores by associating a product code as product identification information with a product name, a category of the product, a price of the product, a size, a description text of the product including how to use the product and usage scenes, and an image of the product, etc., as product information.

[0026] The customer service method DB 122 is a database that stores customer service method information regarding customer service methods corresponding to each of a plurality of products, and accumulates customer service method information based on the remarks of store clerks in the store. The customer service method DB 122 is set to be referable by a large language model described later, similar to the product DB 121. The customer service method DB 122 stores customer service method information in which a plurality of products and products related to each of the plurality of products are associated.

[0027] FIG. 4 is a diagram showing an example of the customer service method DB 122. As shown in FIG. 4, the customer service method DB stores by associating a customer service method ID (Identification) for identifying the customer service method, a customer service method in which at least two products are associated, and a clerk ID for identifying the clerk who proposed the customer service method. Here, it is assumed that the category of at least two of these products is associated with the customer service method in order to associate at least two products, but not limited thereto, and at least two product names or product codes may be included so that at least two of these products are associated.

[0028] The control unit 13 is, for example, a CPU (Central Processing Unit). The control unit 13 functions as an extraction unit 131, a storage control unit 132, a reception unit 133, a first acquisition unit 134, a specification unit 135, a second acquisition unit 136, a generation unit 137, and an output unit 138 by executing a program stored in the storage unit 12.

[0029] [Registration of Customer Service Method Information] First, the function of registering customer service method information will be described. The extraction unit 131 and the storage control unit 132 cooperate to register customer service method information based on the speech of a clerk in the store in the customer service method DB.

[0030] Specifically, the extraction unit 131 acquires text information indicating the speech of a clerk in the store. For example, a microphone is attached to a clerk in the store, and voice information indicating the voice collected by the microphone is converted into text information, and the converted text information is associated with a clerk ID for identifying the clerk wearing the microphone and stored in a storage device (not shown). The extraction unit 131 accesses the storage device to acquire text information indicating the speech of the clerk and the clerk ID. Then, the extraction unit 131 extracts text information corresponding to the customer service method and product identification information, which is information for identifying a product, from the acquired text information.

[0031] For example, the extraction unit 131 includes the acquired text information, classifies the text information into customer service methods, product information, and other information, and inputs a prompt for instructing the large language model to extract one or more product identification information from the text information classified as customer service methods. Then, the extraction unit 131 obtains the text information classified as customer service methods and one or more product identification information classified by the large language model, and extracts the text information corresponding to the customer service method and the product identification information from the acquired text information. Here, the extraction unit 131 may extract the category of the product associated with the product identification information instead of the product identification information.

[0032] The memory control unit 132 stores the text information extracted by the extraction unit 131 in the customer service method DB 122 as customer service method information. Specifically, first, the memory control unit 132 identifies the category of the products handled by the store clerk. For example, the store clerk information associating the store clerk ID of the store clerk with the category of the products handled by the store clerk is stored in the memory unit 12. The memory control unit 132 identifies the category of the products handled by the store clerk by referring to the store clerk information and identifying the category associated with the store clerk ID acquired by the extraction unit 131.

[0033] In addition, the memory control unit 132 identifies the category of the product corresponding to the extracted text information. When a plurality of product identification information is extracted for the text information, the memory control unit 132 identifies the category corresponding to each of the plurality of product identification information. For example, the memory control unit 132 identifies the category included in the product information indicated by the product identification information corresponding to the extracted text information.

[0034] Then, the memory control unit 132 stores the extracted text information in the customer service method DB 122 as customer service method information on the condition that the category of the products handled by the store clerk matches the category of the product corresponding to the extracted text information.

[0035] For example, when a plurality of product identification information is extracted from the text information, the memory control unit 132 associates the plurality of product identification information extracted by the extraction unit 131, the customer service method information, and the store clerk ID with each other and stores them in the customer service method DB 122 on the condition that at least one of the category of the products handled by the store clerk and the category corresponding to each of the plurality of product identification information matches. Here, the memory control unit 132 may associate the category of each of the plurality of products extracted by the extraction unit 131, the customer service method information, and the store clerk ID with each other and store them in the customer service method DB 122. Further, when the customer service method information includes a plurality of product identification information or the category of each of the plurality of products, the memory control unit 132 may associate the customer service method information and the store clerk ID with each other and store them in the customer service method DB 122, thereby associating the product identification information and the customer service method information. By doing so, the information processing device 1 can register the customer service method of a store clerk with rich experience in handling products as the customer service method information in the customer service method DB 122.

[0036] Note that the memory control unit 132 may store the customer service method information in the customer service method DB 122 on the condition that the store clerk is a veteran store clerk who is used to customer service. In this case, for example, the memory unit 12 stores the store clerk ID and the proficiency of the store clerk in customer service in association with each other. The proficiency may be determined based on, for example, the handling period of the product by the user, the evaluation from other users for customer service, and the rank of the position of the user in the store. Here, the evaluation of the customer service method information may be received from other users, and the proficiency may be determined based on the number of good evaluations received, or the proficiency may be determined based on the number of views of the customer service method information.

[0037] Then, when a plurality of product identification information is extracted from the text information, the memory control unit 132 stores the customer service method information in the customer service method DB 122 on the condition that at least one of the category of the products handled by the store clerk and the category corresponding to each of the plurality of product identification information matches and the proficiency of the store clerk is equal to or higher than a predetermined proficiency. By doing so, the information processing device 1 can register only the sophisticated customer service method information in the customer service method DB 122.

[0038] Further, the memory control unit 132 refers to user information in which a user ID as user identification information for identifying each of a plurality of users who have used the store is associated with attribute information indicating the attributes of the users, and specifies the attribute information associated with the user ID of the user who was served by the store clerk in the store.

[0039] In this case, the user ID of the user and the attribute information of the user are associated and stored in the storage unit 12. Also, in a storage device that stores text information converted from the voice collected by the microphone worn by the store clerk, the user ID of the user served by the store clerk is associated with the text information and stored. The user ID may be read, for example, from a point card or a membership card presented by the user when purchasing a product in the store.

[0040] The memory control unit 132 specifies the user ID stored in the storage device in association with the text information acquired by the extraction unit 131, and specifies the attribute information associated with the user ID, thereby specifying the attribute information associated with the user ID of the user who was served by the store clerk in the store. Then, the memory control unit 132 associates the customer service method information extracted from the text information acquired by the extraction unit 131 with the specified attribute information and stores it in the customer service method DB 122. By doing so, the information processing apparatus 1 can store customer service method information with a high probability of being effective for the attribute in association with the attribute of the user who performed the customer service method.

[0041] [Provision of Product Information] Subsequently, a function of providing product information in response to receiving an inquiry about a product from a user will be described. The reception unit 133, the first acquisition unit 134, the specification unit 135, the second acquisition unit 136, the generation unit 137, and the output unit 138 cooperate to output the product information of the product to the user terminal 2 in response to receiving an inquiry about the product from the user terminal 2.

[0042] FIG. 5 is a sequence diagram showing the flow of the process in which the information processing apparatus 1 provides product information. Hereinafter, the function of providing product information will be described with appropriate reference to FIG. 5.

[0043] First, the reception unit 133 receives an inquiry about a product from the user terminal 2 used by the user (S1). For example, the reception unit 133 receives an inquiry about a product from the user terminal 2 via a web page that receives inquiries about products operated by a store or an operator who operates a product information providing service for providing product information.

[0044] The first acquisition unit 134 refers to the product DB 121 that stores product information for each of a plurality of products, and acquires the product information of the product corresponding to the inquiry. For example, the first acquisition unit 134 inputs the inquiry received by the reception unit 133 into a large language model that can refer to the product DB 121 and the customer service method DB 122 (S2).

[0045] For example, when the inquiry received by the reception unit 133 is an inquiry such as "What is a recommended parasol?", the first acquisition unit 134 inputs the inquiry into the large language model. The large language model accesses the product DB 121 to search for product information and acquires the product information corresponding to the inquiry (S3).

[0046] For example, when the inquiry received by the reception unit 133 is "What is a recommended parasol?", the large language model acquires, as the product information corresponding to the inquiry, parasol A as the product name and the description of parasol A, and parasol B as the product name and the description of parasol B. The large language model may further acquire information different from the description included in the product information. The large language model outputs an answer including the acquired product information to the information processing apparatus 1 (S4). The first acquisition unit 134 acquires the product information of the product corresponding to the inquiry by acquiring the answer including the product information corresponding to the inquiry output from the large language model.

[0047] The specific unit 135 refers to the customer service method DB 122, specifies the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit 134, and based on the specified customer service method information, specifies related products related to the product indicated by the product information acquired by the first acquisition unit 134. For example, the specific unit 135 inputs a prompt instructing the large language model to acquire the customer service method information corresponding to the product information acquired by the first acquisition unit 134 (S5).

[0048] Based on the input prompt, the large language model accesses the customer service method DB 122 to search for customer service method information, and acquires the customer service method information corresponding to the product information acquired by the first acquisition unit 134 (S6). The large language model outputs an answer including the acquired customer service method information to the information processing apparatus 1 (S7).

[0049] For example, when the first acquisition unit 134 acquires, as product information, the product name "Sun Umbrella A" and the description of Sun Umbrella A, and the product name "Sun Umbrella B" and the description of Sun Umbrella B, the specific unit 135 generates a prompt such as "Refer to the customer service method DB and acquire the customer service methods corresponding to Sun Umbrella A and Sun Umbrella B.", and inputs the prompt to the large language model. The specific unit 135 acquires, from the large language model, an answer including the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit 134, thereby specifying the customer service method information corresponding to the product indicated by the product information. Then, based on the specified customer service method information, the specific unit 135 specifies related products related to the product indicated by the product information acquired by the first acquisition unit 134 (S8).

[0050] For example, when the first acquisition unit 134 acquires, as product information, the product name "Sun Umbrella A" and the description of Sun Umbrella A, and the product name "Sun Umbrella B" and the description of Sun Umbrella B, the specific unit 135 specifies, as the customer service method information, "When a customer searches for a sun umbrella or sunscreen, also mention a neck cooler as a recommended product." Then, based on the specified customer service method information, the specific unit 135 specifies a neck cooler as a related product of Sun Umbrella A and Sun Umbrella B.

[0051] Here, the specifying unit 135 may specify the clerk ID associated with the specified customer service method information, and associate the specified clerk ID with the usage record indicating that the customer service method information has been used. By doing so, the information processing apparatus 1 can evaluate the clerk indicated by the clerk ID based on the use of the customer service method information.

[0052] Further, when the attribute information indicating the attribute of the user corresponding to the customer service method is associated with the customer service method information in the customer service method DB 122, the specifying unit 135 may specify the attribute of the user using the user terminal 2.

[0053] The specifying unit 135 may specify the attribute of the user by causing the large language model to estimate the attribute of the user based on the conversation history between the large language model and the user using the user terminal 2.

[0054] For example, the storage unit 12 stores the conversation history between the large language model and the user using the user terminal 2 in association with the user ID. The conversation history may include, in addition to the direct conversation history between the large language model and the user, the conversation history between the large language model and the user using the user terminal 2 via the information processing apparatus 1.

[0055] The specifying unit 135 acquires the conversation history between the large language model and the user using the user terminal 2 based on the user ID of the user. Then, the specifying unit 135 generates a prompt including the acquired conversation history, such as "Use the content of the chat between the following AI and the customer to infer, extract, and profile the customer's preferences and characteristics.", and inputs the prompt to the large language model. The large language model estimates the attribute of the user based on the input prompt and outputs it to the information processing apparatus 1. The specifying unit 135 acquires the attribute of the user output from the large language model, thereby obtaining the attribute of the user based on the conversation history between the large language model and the user using the user terminal 2.

[0056] Then, the specifying unit 135 refers to the customer service method DB 122 and specifies the customer service method information corresponding to the attributes of the specified user and the product indicated by the product information acquired by the first acquisition unit 134. By doing so, the information processing apparatus 1 can specify effective customer service method information for the user.

[0057] The second acquisition unit 136 refers to the product information and acquires the product information corresponding to the related product specified by the specifying unit 135. For example, the second acquisition unit 136 inputs a prompt for instructing the acquisition of the product information corresponding to the related product specified by the specifying unit 135 into the large language model (S9). Based on the input prompt, the large language model accesses the product DB 121, searches for the product information, and acquires the product information corresponding to the prompt, that is, the product information corresponding to the related product (S10). The large language model outputs an answer including the acquired product information to the information processing apparatus 1 (S11). The second acquisition unit 136 acquires the product information corresponding to the related product by acquiring the answer including the product information output from the large language model.

[0058] When the specifying unit 135 specifies a neck cooler as the related product, the second acquisition unit 136 generates a prompt such as "Please acquire the product information of the neck cooler." and inputs the prompt into the large language model. Then, the second acquisition unit 136 acquires the product information of the neck cooler as the product information corresponding to the related product from the large language model.

[0059] The generation unit 137 inputs a prompt instructing the large language model to create a response sentence corresponding to a user inquiry, including the product information obtained by the first acquisition unit 134 and the related product information obtained by the second acquisition unit 136, which prompt the purchase of the product indicated by the product information obtained by the first acquisition unit 134 and the related product indicated by the product information obtained by the second acquisition unit 136 (S12). The large language model generates a response sentence corresponding to the user inquiry, including the product information obtained by the first acquisition unit 134 and the product information obtained by the second acquisition unit 136, and outputs the response sentence to the information processing apparatus 1 (S13). The second acquisition unit 136 acquires the response sentence output from the large language model.

[0060] For example, the generation unit 137 generates a prompt such as "Based on the description of umbrella A, the description of umbrella B, and the description of the neck cooler, create a sentence encouraging the purchase of umbrella A or umbrella B together with the neck cooler.", and inputs the prompt to the large language model. Then, the second acquisition unit 136 acquires from the large language model a response sentence corresponding to the user inquiry.

[0061] The output unit 138 outputs the product information obtained by the first acquisition unit 134 and the product information obtained by the second acquisition unit 136 to the user terminal 2. Specifically, the output unit 138 outputs to the user terminal 2 a response sentence generated by the generation unit 137 and including the product information obtained by the first acquisition unit 134 and the product information obtained by the second acquisition unit 136 (S14).

[0062] [Effect by the information processing apparatus 1] As described above, when the information processing apparatus 1 according to the present embodiment receives an inquiry about a product from the user terminal 2, it refers to the product DB 121, acquires product information of the product corresponding to the inquiry, and stores in the customer service method DB 122 customer service method information regarding the customer service method corresponding to each of the plurality of products, which associates a plurality of products with products related to each of the plurality of products, and refers to the customer service method DB 122 to identify the customer service method information corresponding to the product indicated by the acquired product information. Then, based on the identified customer service method information, the information processing apparatus 1 identifies related products related to the product indicated by the product information, refers to the product DB 121 to acquire product information corresponding to the related products, and outputs the product information of the product corresponding to the inquiry and the product information of the related products to the user terminal 2. By doing so, the information processing apparatus 1 can enhance the persuasive power when proposing products.

[0063] In addition, according to the present invention, it becomes possible to contribute to Goal 9, "Build the infrastructure for industry and technological innovation," of the Sustainable Development Goals (SDGs) led by the United Nations.

[0064] As described above, the present invention has been described using embodiments. However, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist thereof. For example, all or part of the device can be configured by being functionally or physically distributed and integrated in any unit. Also, new embodiments resulting from any combination of a plurality of embodiments are included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination have the effects of the original embodiments combined.

Explanation of Reference Numerals

[0065] 1 Information processing apparatus 2 User terminal 11 Communication unit 12 Storage unit 13 Control unit 131 Extraction unit 132 Storage control unit 133 Reception unit 134 First acquisition unit 135 Identification unit 136 Second acquisition unit 137 Generation unit 138 Output unit

Claims

An extraction unit that obtains text information indicating the speech of a store clerk in a store, generates a prompt that includes the obtained text information and instructs to classify the text information into text information indicating a customer service method that associates a product with related products and is different from the text information indicating the customer service method, inputs the prompt into a large language model, and extracts text information corresponding to the customer service method from the obtained text information by obtaining the text information classified into the customer service method output from the large language model. A storage control unit that refers to store clerk information associating store clerk identification information for identifying the store clerk with the category of products or services handled by the store clerk, specifies the category of products or services associated with the store clerk identification information of the store clerk as the category of products or services handled by the store clerk, extracts product identification information included in the extracted text information and pre-associated with the category of products or services, specifies the category of products or services associated with the extracted product identification information, and stores the text information extracted by the extraction unit in a customer service method database as customer service method information on the condition that the category of products or services handled by the store clerk matches the category of products or services corresponding to the extracted text information. A reception unit that receives an inquiry regarding the product, including text information indicating the product, from a user terminal used by the user. A first acquisition unit that refers to a product database storing product information, which is text information regarding each of a plurality of products and is associated with the text information indicating the product, and acquires product information of a product corresponding to the text information included in the inquiry. A specifying unit that refers to the customer service method database storing text information indicating a customer service method corresponding to each of a plurality of products, which associates a plurality of products with related products respectively, specifies customer service method information including the text information indicating the product indicated by the product information acquired by the first acquisition unit, and specifies related products indicated by the text information of products related to the product indicated by the product information acquired by the first acquisition unit included in the specified customer service method information. A second acquisition unit that refers to the product database and acquires product information corresponding to the related products specified by the specifying unit; An output unit that outputs the product information acquired by the first acquisition unit and the product information acquired by the second acquisition unit to the user terminal; An information processing apparatus having the above.

2. The first acquisition unit refers to the product database to search for product information, inputs the inquiry to a large language model that can refer to the customer service method database, and obtains an answer including the product information retrieved by the large language model for the inquiry from the large language model, thereby acquiring product information of the product corresponding to the inquiry. The specifying unit generates a prompt including text information indicating the product corresponding to the product information acquired by the first acquisition unit and instructing the acquisition of customer service method information corresponding to the product, inputs the generated prompt to the large language model, and obtains from the large language model an answer including the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit, thereby specifying the customer service method information corresponding to the product indicated by the product information acquired by the first acquisition unit. The second acquisition unit generates a prompt including text information indicating the related products specified by the specifying unit and instructing the acquisition of product information corresponding to the related products, inputs the generated prompt to the large language model, and obtains an answer including the product information output from the large language model, thereby acquiring the product information corresponding to the related products. The information processing apparatus according to claim 1.

3. A generation unit that generates a prompt for instructing the creation of a response sentence corresponding to the inquiry, including text information indicating the product corresponding to the product information acquired by the first acquisition unit and text information indicating the related products corresponding to the product information acquired by the second acquisition unit, which prompts the purchase of the product indicated by the product information acquired by the first acquisition unit and the related products indicated by the product information acquired by the second acquisition unit for the large language model, inputs the generated prompt to the large language model, and obtains the response sentence from the large language model, thereby generating the response sentence. The output unit outputs the response sentence generated by the generation unit. The information processing apparatus according to claim 2.

4. The customer service method information is associated with salesperson identification information for identifying the salesperson who provided customer service corresponding to the customer service method. The specifying unit specifies the salesperson identification information associated with the specified customer service method information, and associates a usage record indicating that the customer service method information has been used with the specified salesperson identification information. The information processing apparatus according to claim 1.

5. The extraction unit generates a prompt that includes the acquired text information and instructs to extract product identification information, which is text information for identifying a product, from the text information, inputs the prompt into a large language model, and acquires the product identification information output from the large language model, thereby extracting information for identifying a product from the acquired text information. The storage control unit associates the information for identifying the product extracted by the extraction unit with the text information corresponding to the customer service method and stores it in the customer service method database as the customer service method information. The information processing apparatus according to claim 1.

6. The storage control unit refers to user information in which user identification information for identifying each of a plurality of users is associated with attribute information indicating the attributes of the users, specifies the attribute information associated with the user identification information of the user served by the salesperson in the store, and associates the customer service method information with the specified attribute information and stores it in the customer service method database. The information processing apparatus according to claim 1.

7. Executed by a computer, acquiring text information indicating the utterances of salespersons in a store, generating a prompt that includes the acquired text information and instructs to classify the text information into text information indicating a customer service method that associates a product with related products and text information different from the text information indicating the customer service method, inputting the prompt into a large language model, and acquiring the text information classified into the customer service method output from the large language model, thereby extracting text information corresponding to the customer service method from the acquired text information. Referring to the store employee information that associates the store employee identification information for identifying the store employee with the category of products or services handled by the store employee, specifying the category of products or services associated with the store employee identification information of the store employee as the category of products or services handled by the store employee, and extracting product identification information included in the extracted text information, which is product identification information pre-associated with the category of products or services, specifying the category of products or services associated with the extracted product identification information, and storing the extracted text information in the customer service method database as customer service method information on the condition that the category of products or services handled by the store employee matches the category of products or services corresponding to the extracted text information; Receiving an inquiry regarding the product, including text information indicating the product, from a user terminal used by the user; Referring to a product database that stores product information, which is text information regarding each of a plurality of products and is associated with the text information indicating the product, and obtaining the product information of the product corresponding to the text information included in the inquiry; Referring to the customer service method database that stores customer service method information, which is text information indicating the customer service method corresponding to each of a plurality of products and associates each of the plurality of products with the products related to each of the plurality of products, specifying the customer service method information including the text information indicating the product indicated by the obtained product information, and specifying the related products indicated by the text information of the products related to the product indicated by the obtained product information included in the specified customer service method information; Referring to the product database and obtaining the product information corresponding to the specified related products; Outputting the product information of the product corresponding to the obtained inquiry and the product information of the obtained related products to the user terminal; An information processing method having the above steps.

Images