Information processing method, program, and information processing device

The method uses a generative model to guide users to product pages with relevant information and cashback incentives, addressing the challenge of ineffective navigation in interactive chat systems.

JP2024157275APending Publication Date: 2024-11-07SOCIALGOOD INC
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023071537
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-25
Publication Date
2024-11-07

AI Technical Summary

Technical Problem

Existing systems fail to effectively guide users to product purchase pages or seller web pages through interactive chat systems.

Method used

An information processing method that utilizes a generative model trained to generate response sentences, incorporating product information and links to purchase pages, and provides cashback incentives, guiding users through a chat interface.

Benefits of technology

Users are suitably guided to product purchase pages or seller web pages with relevant product information and cashback incentives, enhancing the purchasing experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024157275000001_ABST
    Figure 2024157275000001_ABST
Patent Text Reader

Abstract

To provide an information processing method, and the like, capable of properly leading a user to a commodity purchase page or a web page of a seller of the commodity.SOLUTION: An information processing method causes a computer to execute the processes of: acquiring an input sentence; acquiring a response sentence generated by inputting the acquired input sentence to a trained model which has been trained to generate a response sentence when an input sentence is input; and outputting commodity information including a link 51 to a commodity purchase page or a Web page of a seller of the commodity, together with the response sentence.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

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

[0002] As a system for presenting product information to a user, a system has been proposed that has a dialogue with a user in a chat format and presents product information according to the content of the dialogue. For example, Patent Document 1 discloses a product information providing system that sequentially reads out a plurality of types of dialogue templates according to a scenario to generate dialogue text, and when the user responds affirmatively to the dialogue text, generates search keywords related to the dialogue text, searches for products using the generated search keywords, and presents the searched products to the user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2014-167722 A Summary of the Invention [Problem to be solved by the invention]

[0004] In one aspect, an object of the present invention is to provide an information processing method etc. that can suitably guide a user to a product purchase page or a web page of a product seller. [Means for solving the problem]

[0005] In one aspect, the information processing method includes a process in which a computer executes a process of acquiring an input sentence, inputting the acquired input sentence into a model that has been trained to generate a response sentence when an input sentence is input, thereby acquiring a response sentence generated by the input sentence, and outputting product information including a link to a product purchase page or a web page of the product seller together with the response sentence in accordance with the response sentence. Effect of the Invention

[0006] In one aspect, the user can be conveniently directed to a product purchase page or a web page of the product seller. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of the configuration of an information processing system. [Diagram 2] FIG. 2 is a block diagram showing an example of the configuration of a server. [Diagram 3] 1 is an explanatory diagram showing an example of a record layout of a product DB, a user DB, and an affiliated store DB. [Figure 4] FIG. 2 is a block diagram showing a configuration example of a terminal. [Diagram 5] FIG. 13 is an explanatory diagram illustrating an example of a chat screen. [Figure 6] 11 is a flowchart illustrating an example of a processing procedure executed by a server. [Figure 7] FIG. 11 is an explanatory diagram showing an overview of a second embodiment. [Figure 8] FIG. 11 is an explanatory diagram showing a response sentence depending on whether or not a command sentence is present; [Figure 9] 11 is a flowchart illustrating an example of a processing procedure executed by a server according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. (Embodiment 1) 1 is an explanatory diagram showing a configuration example of an information processing system. In this embodiment, a system that uses a generation model 30 constructed by machine learning to have a dialogue with a user and that inserts product information including a link to a product purchase page or a web page of a product seller into a response sentence generated by the generation model 30 in response to a sentence input by the user, and outputs the response sentence. The information processing system includes an information processing device 1, terminals 2, 2, 2, . . . and a generation server 3. Each device is communicatively connected to a network N such as the Internet.

[0009] The information processing device 1 is an information processing device capable of various information processing and transmitting and receiving information, such as a server computer, a personal computer, etc. In this embodiment, the information processing device 1 is assumed to be a server computer, and will be referred to as server 1 for simplicity in the following. As described later, the server 1 provides a cashback service in which a user purchases a product at an affiliated store (e.g., an EC (Electronic Commerce) site) and is given a virtual currency (Cryptocurrency) in an amount corresponding to the purchase amount of the product. In this embodiment, the server 1 provides a chat function that allows the user to interact with a chatbot as one function of the cashback service. Then, depending on the content of the response text from the chatbot, when the response text includes, for example, the name of a product that can be purchased at an affiliated store, the server 1 inserts product information including a link to the purchase page of the product or a web page (e.g., the top page of an EC site) of the seller (affiliated store) selling the product, the purchase amount of the product, and a cashback rate for the virtual currency given into the response text and outputs the response text to the user.

[0010] Terminal 2 is a terminal device used by a user, such as a smartphone, a tablet terminal, a personal computer, etc. Terminal 2 displays a chat screen (see FIG. 5) described later, accepts input of an input sentence from a user, and displays a response sentence to the input sentence.

[0011] The generation server 3 is a server computer that generates a response sentence to a user's input sentence by using a generation model 30 constructed by machine learning. The generation server 3 receives an input sentence from the server 1 through an API (Application Programming Interface) provided by the generation server 3, generates a response sentence by inputting the input sentence into the generation model 30, and outputs the generated response sentence to the server 1.

[0012] The generative model 30 is a machine learning model that has learned predetermined training data, and is, for example, a Generative Pre-trained Transformer (GPT), which is a type of Transformer. In this embodiment, ChatGPT provided by OpenAI, Inc. is used as the generative model 30.

[0013] In this embodiment, ChatGPT is used, but the generation model 30 may be a model other than GPT. Also, the server 1 does not necessarily need to acquire the response sentence through an external system (generation server 3), and the server 1 may generate the response sentence by itself.

[0014] 2 is a block diagram showing an example of the configuration of the server 1. The server 1 includes a control unit 11, a main storage unit 12, a communication unit 13, and an auxiliary storage unit . The control unit 11 has one or more arithmetic processing devices such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), etc., and performs various information processing, control processing, etc. by reading and executing a program P1 stored in the auxiliary storage unit 14. The main storage unit 12 is a temporary storage area such as an SRAM (Static Random Access Memory) or a DRAM (Dynamic Random Access Memory), and temporarily stores data required for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing processing related to communication, and transmits and receives information to and from the outside.

[0015] The auxiliary storage unit 14 is a non-volatile storage area such as a large capacity memory or a hard disk, and stores a program P1 (program product) and other data necessary for the control unit 11 to execute processing. The auxiliary storage unit 14 also stores a product DB 141, a user DB 142, and an affiliated store DB 143. The product DB 141 is a database that stores information on products provided by affiliated stores. The user DB 142 is a database that stores information on users of the system. The affiliated store DB 143 is a database that stores information on affiliated stores.

[0016] The auxiliary storage unit 14 may be an external storage device connected to the server 1. The server 1 may be a multi-computer consisting of a plurality of computers, or may be a virtual machine virtually constructed by software.

[0017] In the present embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit for accepting operation input, a display unit for displaying images, etc. The server 1 may also include a reading unit for reading a portable storage medium 1a such as a CD (Compact Disk)-ROM or a DVD (Digital Versatile Disc)-ROM, and may read and execute the program P1 from the portable storage medium 1a.

[0018] FIG. 3 is an explanatory diagram showing an example of the record layout of the product DB 141, the user DB 142, and the affiliated store DB 143. As shown in FIG. Product DB141 includes a product ID column, a product name column, a category column, a member store providing the product, a price column, a purchase page column, and a keyword column. The product ID column stores a product ID for identifying each product. The product name column, category column, member store providing the product, price column, purchase page column, and keyword column each store, in association with a product ID, the product name, product category, ID of the member store providing the product, product purchase price, link information (address) of the product purchase page at the member store (EC site, etc.), and keywords. Keywords will be described later.

[0019] The user DB 142 includes a user ID column, a name column, an attribute column, a purchase history column, an owned coin column, and a wallet address column. The user ID column stores a user ID for identifying each user. The name column, attribute column, purchase history column, owned coin column, and wallet address column store, in association with the user ID, the user's name, attribute (age, gender, etc.), product purchase history at affiliated stores, the amount of virtual currency (coins) currently held by the user, and the user's wallet address.

[0020] The affiliated store DB143 includes an affiliated store ID column, an affiliated store name column, an affiliated store page column, and a cashback rate column. The affiliated store ID column stores an affiliated store ID for identifying each affiliated store. The affiliated store name column, the affiliated store page column, and the cashback rate column each store, in association with the affiliated store ID, the name of the affiliated store, the address of the web page of the affiliated store (the seller of the product), and the cashback rate that is returned to the user in the form of virtual currency when a product is purchased at the affiliated store.

[0021] In this embodiment, the cashback rate is assumed to be determined for each affiliated store, but the cashback rate may be determined for each product, for example.

[0022] 4 is a block diagram showing an example of the configuration of the terminal 2. The terminal 2 includes a control unit 21, a main memory unit 22, a communication unit 23, a display unit 24, an input unit 25, and an auxiliary memory unit . The control unit 21 has one or more processors such as CPUs, and performs various information processing by reading and executing the program P2 stored in the auxiliary storage unit 26. The main storage unit 22 is a temporary storage area such as RAM, and temporarily stores data necessary for the control unit 21 to execute arithmetic processing. The communication unit 23 is a communication module for performing processing related to communication, and transmits and receives information to and from the outside. The display unit 24 is a display screen such as a liquid crystal display, and displays images. The input unit 25 is an operation interface such as a touch panel, and accepts operation input from the user. The auxiliary storage unit 26 is a non-volatile storage area such as a hard disk, and stores the program P2 (program product) and other data necessary for the control unit 21 to execute processing.

[0023] The terminal 2 may include a reading unit for reading a portable storage medium 2a such as a CD-ROM, and may read and execute the program P2 from the portable storage medium 2a.

[0024] Fig. 5 is an explanatory diagram showing an example of a chat screen. Fig. 5 illustrates a state in which a response sentence generated by the generative model 30 is displayed in response to an input sentence "I want a camera that can be used at the sports day" input by a user.

[0025] Before explaining with reference to FIG. 5, the cashback service provided by the server 1 will be described. As described above, the server 1 provides a cashback service in which, when a user purchases a product at an affiliated store, the server 1 grants the user an amount of virtual currency corresponding to the purchase price of the product. That is, when a user purchases a product at an affiliated store, the server 1 transfers to the user's wallet address an amount of virtual currency corresponding to the cashback amount obtained by multiplying the purchase price of the product by the cashback rate determined in the affiliated store DB 143. The virtual currency granted to the user is funded by the fees collected from the affiliated store.

[0026] In this embodiment, the cashback will be explained as being provided in the form of virtual currency, but the cashback may be provided in the form of other electronic media (for example, points).

[0027] To realize this service, a dedicated application (hereinafter referred to as "app") is installed on the user's device 2. The user can use the links of each affiliated store and individual product displayed on the app to navigate to the affiliated store's product purchase page and shop. In addition, the user can check the current balance of virtual currency on the app.

[0028] In this embodiment, as one function of the application, a chat service that allows a user to interact with a chatbot is provided. Fig. 5 shows an example of a screen during chat. The server 1 accepts an input sentence on the chat screen in Fig. 5, and transmits the input sentence to the generation server 3. The server 1 then obtains from the generation server 3 a response sentence that has been generated by inputting the input sentence into the generation model 30, and outputs the response sentence to the chat screen.

[0029] If the response text contains words or phrases related to a product provided by an affiliated store, the server 1 outputs product information including a link 51 to a product purchase page or a web page (e.g., the top page of an EC site) of the product seller (affiliated store) together with the response text. In Fig. 5, in addition to the link 51, the product information includes the purchase price of the product and the cashback rate to be returned to the user by granting virtual currency.

[0030] For example, product DB 141 stores product-related keywords in association with product IDs. The keywords are, for example, product names, but may also be words expressing the type of product ("Single-lens reflex camera" if the product is a single-lens reflex camera), manufacturer names, etc. Server 1 refers to product DB 141 and detects product-related keywords from the response sentence. In FIG. 5, the product names "Camera A", "Camera B", and "Camera B" detected as keywords are shown in bold.

[0031] When a keyword is detected, the server 1 reads out product information corresponding to the detected keyword from each database. Specifically, the server 1 reads out a link 51 (address) to the product purchase page or the web page of the product seller (affiliated store), the purchase price of the product, and the cashback rate determined for each affiliated store.

[0032] The server 1 inserts the product information at a position corresponding to the keyword detected from the response text. For example, as shown in Fig. 5, the server 1 inserts a link 51 immediately after the appearance of the keyword (product name) and also inserts text indicating the purchase price of the product and the cashback rate (cashback information). The server 1 outputs the response text with the product information inserted to the terminal 2 and causes it to be displayed.

[0033] In this embodiment, the cashback rate is displayed together with the link 51, the purchase amount of the product, etc., but the present embodiment is not limited to this. For example, the server 1 may calculate the cashback amount by multiplying the purchase amount of the product by the cashback rate, and output the cashback amount on the chat screen. In other words, the server 1 only needs to be able to display, on the chat screen, the cashback information that is returned to the user by granting virtual currency, and the cashback information is not limited to the cashback rate.

[0034] When the user transitions to a product purchase page or a web page (not shown) in response to an operation input to the link 51 and accepts a product purchase application, the server 1 grants the user an amount of virtual currency corresponding to the purchase price of the product. That is, the server 1 generates a transaction for transferring an amount of virtual currency corresponding to the cashback amount to the user's wallet address, and outputs (broadcasts) the transaction to the blockchain network.

[0035] As described above, according to this embodiment, by inserting the link 51 to the product purchase page into the response sentence generated by the generative model 30, it is possible to conveniently guide the user to the product purchase page.

[0036] In addition, if keywords for multiple products are detected from the response sentence, i.e., if the response sentence contains descriptions of multiple products, the server 1 may rearrange the order of the descriptions of each product depending on the purchase amount or cashback amount of each product.

[0037] For example, the server 1 extracts a group of sentences from the appearance of one keyword to the appearance of the next keyword as one group of sentences according to the position where each keyword is detected, thereby dividing the response sentence into a plurality of groups of sentences corresponding to each product. In the example of FIG. 5, the group of sentences from the keyword "Camera A" to the appearance of the next keyword "Camera B" is extracted as one group of sentences, "Camera A / SLR camera, equipped with an AF system suitable for sports photography." Then, the server 1 generates a response sentence by rearranging each of the divided groups of sentences in ascending order of the purchase amount of each product or in descending order of the cashback amount. For example, when rearranging in descending order of the purchase amount, the order is "Camera B", "Camera A", and "Camera C". It is needless to say that product information such as the link 51 is inserted immediately after the keyword. This allows information on each product to be presented to the user in descending order of the price of the product or the amount of the cashback.

[0038] 6 is a flowchart showing an example of a processing procedure executed by the server 1. The processing contents executed by the server 1 will be described with reference to FIG. The control unit 11 of the server 1 acquires an input sentence entered by a user from the terminal 2 (step S11). The control unit 11 transmits the acquired input sentence to the generation server 3, and acquires from the generation server 3 a response sentence that has been generated by inputting the input sentence into the generation model 30 (step S12).

[0039] The control unit 11 refers to the product DB 141 and determines whether or not a keyword related to the product is detected from the response sentence (step S13). If it is determined that the keyword is not detected (S13: NO), the control unit 11 outputs the response sentence acquired from the generation server 3 to the terminal 2 as is (step S14), and ends the series of processes.

[0040] When it is determined that a keyword has been detected (S13: YES), the control unit 11 reads out from each database a link 51 to a product purchase page or a product seller's web page corresponding to the keyword, product purchase amount, cashback rate, and other product information (step S15). The control unit 11 outputs the product information including the link 51 to the product purchase page or web page together with the response message to the terminal 2 (step S16). Specifically, the control unit 11 inserts the product information into a position corresponding to the keyword detected from the response message, and outputs the product information to the terminal 2.

[0041] The control unit 11 transitions to a product purchase page or a product seller's web page based on the link 51 inserted as product information and determines whether the product purchase application has been accepted (step S17). If it is determined that the product purchase application has been accepted (S17: YES), the control unit 11 grants the user an amount of virtual currency according to the purchase price of the product (step S18). After executing the process of step S18, or if the answer is NO in step S17, the control unit 11 ends the series of processes.

[0042] As described above, according to the first embodiment, it is possible to suitably guide the user to a product purchase page or a web page of a product seller.

[0043] (Embodiment 2) In this embodiment, a form will be described in which the accuracy of a response sentence is improved by inputting information such as a user's product purchase history and product browsing history into the generative model 30. Note that the same reference numerals will be used to designate the same contents as in the first embodiment, and the description thereof will be omitted.

[0044] Fig. 7 is an explanatory diagram showing an overview of the second embodiment. The overview of the present embodiment will be described with reference to Fig. 7.

[0045] 7 illustrates a state in which a response sentence is generated by inputting information such as the user's product purchase history and browsing history, and an instruction sentence for indicating the focus of the response sentence, in addition to an input sentence input by the user, into the generation model 30. In this embodiment, the server 1 transmits the input sentence, information such as the product purchase history, and the instruction sentence as a series of text to the generation server 3, thereby generating a response sentence for the text.

[0046] Other information to be input to the generative model 30 besides the input sentence includes the user's product purchase history, product browsing history at affiliated stores (EC sites), past dialogue history on the chat screen (e.g., product names that appeared in past dialogues), user attributes, product attributes (categories) detected as keywords, or product seller attributes (e.g., types of EC sites). When an input sentence is accepted on the chat screen, the server 1 reads out this information from each database or obtains it from the outside (affiliated stores). The server 1 then connects text listing, for example, product names that the user has previously purchased or viewed, product names that appeared in past chats, the user's age, gender, and the like, to the end of the input sentence to create a series of texts. By adding this information to the input sentence, the generative model 30 can generate a response sentence while referring to the user's product purchasing tendencies and attributes.

[0047] Note that product purchase history and the like are examples of user information, and the present embodiment is not limited thereto. For example, the user information may include information regarding the user's holding or use of virtual currency. Information regarding the holding or use of virtual currency is, for example, the amount of virtual currency held by the user, the holding period, and the frequency with which the user uses virtual currency. By including this information in the user information, it is possible to identify whether the user is a frequent user of the cashback service provided by this system, and to present, for example, products with a high cashback rate.

[0048] The instruction sentence is text for indicating the focus of the response sentence, and is information for controlling the generation of the response sentence by the generative model 30. For example, the server 1 generates a series of text to be input to the generative model 30 by connecting the instruction sentence to the end of the text following the input sentence and user information.

[0049] The instruction may be specified by the user, or the server 1 may automatically specify the instruction. Explaining based on the example of Fig. 5, for example, if the input sentence is "I want a camera that can be used at the athletic meet", it is conceivable that "product", "low price", "sports", etc. are specified as the instruction. In this embodiment, in order to include information on products sold at affiliated stores in the response sentence, for example, the word "product" is specified as the instruction.

[0050] Fig. 8 is an explanatory diagram showing response sentences depending on whether or not a command statement is added. Fig. 8 shows an example of a response sentence when a command statement is not added to an input sentence (Fig. 8A) and a response sentence when a command statement is added to an input sentence (Fig. 8B).

[0051] For example, as shown in FIG. 8A, when the user inputs only the input sentence "I want a camera that can be used at the sports day" to the generative model 30, the generative model 30 may generate a response sentence that lists the camera functions required at the sports day, rather than introducing a product. In contrast, as shown in FIG. 8B, when the user inputs the instruction "product" to the generative model 30 in addition to the input sentence "I want a camera that can be used at the sports day", the generative model 30 generates a response sentence that focuses on the product. In this way, by adding an instruction to the text input to the generative model 30, an appropriate response sentence can be generated.

[0052] For example, the server 1 uploads a large amount of training data to the generation server 3 in advance and causes the generation model 30 to learn the data, thereby fine-tuning the generation model 30. That is, the server 1 uploads training data in which correct response sentences are associated with text consisting of an input sentence, information such as a product purchase history, and an instruction sentence, to the generation model 30, and causes the generation model 30 to learn the data.

[0053] When actually accepting an input sentence from a user and outputting a response sentence, the server 1 transmits the input sentence together with information such as product purchase history and the instruction sentence to the generation server 3. The server 1 then obtains from the generation server 3 the response sentence that has been generated by inputting the input sentence, information such as product purchase history and the instruction sentence into a fine-tuned generative model 30. The subsequent processing is the same as in the first embodiment, and the server 1 inserts product information including a link 51 to a product purchase page into the response sentence and outputs it to the chat screen.

[0054] 9 is a flowchart showing an example of a processing procedure executed by the server 1 according to the embodiment 2. After acquiring an input sentence from the terminal 2 (step S11), the server 1 executes the following processing. The control unit 11 of the server 1 acquires information including the user's product purchase history, product browsing history, dialogue history, user attributes, product attributes, or product seller attributes (step S201). The control unit 11 also acquires an instruction sentence for indicating the focus of the response sentence (step S202). The control unit 11 transmits the input sentence, information such as the product purchase history, and the instruction sentence as a series of text to the generation server 3, and acquires from the generation server 3 a response sentence generated by inputting the text into the generation model 30 (step S203). The control unit 11 transitions the process to step S13.

[0055] As described above, according to the second embodiment, by inputting information such as a product purchase history and an instruction sentence to the generative model 30 in addition to the input sentence, it is possible to generate a more appropriate response sentence.

[0056] The embodiments disclosed herein are illustrative in all respects and should not be considered as limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the scope and meaning equivalent to the claims.

[0057] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations regardless of the citation format. Furthermore, the claims use a format in which a claim cites two or more other claims (multi-claim format), but this is not limited to this. They may also be written using a format in which a multiple claim cites at least one other multiple claim (multi-multi claim). [Explanation of symbols]

[0058] 1. Server (information processing device) 11 Control section 12 Main memory 13. Communications Department 14 Auxiliary storage P1 Program 141 Product DB 142 User DB 143 Member store DB 2. Terminal 21 Control section 22 Main memory 23 Communications Department 24 Display section 25 Input section 26 Auxiliary storage P2 Program 3 Generation Server 30 Generative Model

Claims

1. Get the input sentence, A model that has been trained to generate a response sentence when an input sentence is inputted is provided with a response sentence generated by inputting the input sentence obtained, In response to the response message, product information including a link to a product purchase page or a web page of the product seller is output together with the response message. An information processing method in which processing is performed by a computer.

2. Detecting keywords related to the product from the response sentence; When the keyword is detected, the response text is output in which the product information on the product corresponding to the keyword is inserted at a position corresponding to the keyword. The information processing method according to claim 1 .

3. When the user is redirected to the product purchase page or the Web page based on the link and a product purchase application is accepted, virtual currency is awarded to the user in an amount corresponding to the purchase amount of the product. The information processing method according to claim 1 .

4. Outputting cashback information to be returned to the user by providing the virtual currency together with the response text. The information processing method according to claim 3.

5. Acquire information including a user's product purchase history, product browsing history, interaction history, user attributes, product attributes, or product seller attributes; The response sentence generated by inputting the input sentence and the information into the model is obtained. The information processing method according to claim 1 .

6. obtaining a directive for indicating a focus of the response sentence; Obtaining the response sentence generated by inputting the input sentence, the information, and the instruction sentence into the model. The information processing method according to claim 5.

7. Get the input sentence, A model that has been trained to generate a response sentence when an input sentence is inputted is provided with a response sentence generated by inputting the input sentence obtained, In response to the response message, product information including a link to a product purchase page or a web page of the product seller is output together with the response message. A program that causes a computer to carry out processing.

8. An information processing device including a control unit, The control unit: Get the input sentence, A model that has been trained to generate a response sentence when an input sentence is inputted is provided with a response sentence generated by inputting the input sentence obtained, In response to the response message, product information including a link to a product purchase page or a web page of the product seller is output together with the response message. Information processing device.

Citation Information

Patent Citations

  • Merchandise information providing system and program for providing merchandise information

    JP2014167722A