Information processing system, information processing device, program and information processing method

Abstract

To provide an information processing system, an information processing device, a program, and an information processing method that realize excellent matching between an investor and an advisor.SOLUTION: In an information processing system, an information processing device executes: acceptance processing for accepting responses for preset questions related to investment from an investor as investor information; provision processing for providing the investor information accepted in the acceptance processing to a language model; and scoring processing for collating an investor tag that is generated by the language model on the basis of the investor information provided in the provision processing and is related to requests of the investor for an advisor who provides advice related to investment to the investor with an advisor tag set on the basis of attributes of the advisor so as to assign a score.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing device, a program, and an information processing method, and in particular to an information processing system, an information processing device, a program, and an information processing method that executes information processing for matching investors with advisors. [Background technology]

[0002] For example, there are technologies that can match various needs, such as matching the needs of those who want to take over a business and those who want to receive business succession, those who want to hire a job and those who are looking for work, or those who want to get married.

[0003] Patent Document 1 discloses a matching technology that realizes a service that matches lender users with borrower users based on financial asset information acquired from users. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2023-33547 Summary of the Invention [Problem to be solved by the invention]

[0005] Incidentally, when making an investment in, for example, financial assets or real estate, advice from an advisor with specialized knowledge and experience regarding investments and assets may be useful in reducing the risk associated with the investment.

[0006] On the other hand, there is a concern that investors, especially those who are inexperienced in investing, may not be able to select the most suitable advisor because they may not be clear about which type of advisor is suitable for them.

[0007] The present invention has been made in consideration of the above circumstances, and has an objective of providing an information processing system, an information processing device, a program, and an information processing method that can achieve good matching between investors and advisors. [Means for solving the problem]

[0008] In order to achieve the above-mentioned object, the information processing system of the present invention executes a reception process for receiving answers to predetermined questions regarding investments from investors as investor information, a provision process for providing the investor information received in the reception process to a language model, and a scoring process for matching investor tags related to the investor's requests to advisors who provide investment advice to the investor, which are generated in the language model based on the investor information provided in the provision process, with advisor tags set based on the attributes of the advisors, and assigning scores.

[0009] According to this, a score is assigned by comparing an investor tag relating to an investor's requests to an advisor with an advisor tag relating to the attributes of the advisor. Therefore, for example, by performing a matching process based on the assigned score, good matching between investors and advisors can be achieved.

[0010] This information processing system executes a matching process that matches investors and advisors based on scores assigned in a scoring process, and the scoring process assigns a score based on the degree of match between an investor tag and an advisor tag.

[0011] Incidentally, the investor information may be composed of audio information or text information.

[0012] In order to achieve the above-mentioned object, the information processing device of the present invention is an information processing device having a processor and a memory in which a program is stored, and by executing the program by the processor, it performs a reception process of receiving answers to predetermined questions regarding investments from investors as investor information, a provision process of providing the investor information received in the reception process to a language model, and a scoring process of matching investor tags related to the investor's requests to advisors who provide investment advice to the investor, which are generated in the language model based on the investor information provided in the provision process, with advisor tags set based on the attributes of the advisor, and assigning a score.

[0013] The program of the present invention for achieving the above-mentioned object causes an information processing device implemented by a computer to execute a reception process for receiving answers to predetermined questions regarding investments from investors as investor information, a provision process for providing the investor information received in the reception process to a language model, and a scoring process for matching investor tags related to the investor's requests to advisors who provide investment advice to the investor, which are generated in the language model based on the investor information provided in the provision process, with advisor tags set based on the attributes of the advisor, and assigning a score.

[0014] The information processing method of the present invention for achieving the above-mentioned object involves an information processing device implemented by a computer that executes a reception process in which answers to preset questions regarding investments are received from investors as investor information, a provision process in which the investor information received in the reception process is provided to a language model, and a scoring process in which investor tags related to the investor's requests to advisors who provide investment advice to the investor, which are generated in the language model based on the investor information provided in the provision process, are compared with advisor tags set based on the attributes of the advisor to assign a score. Effect of the Invention

[0015] According to the present invention, good matching between investors and advisors can be achieved. [Brief description of the drawings]

[0016] [Figure 1] 1 is a block diagram illustrating an outline of a configuration of an information processing system according to an embodiment of the present invention. [Diagram 2] FIG. 1 is a block diagram for explaining an outline of the configuration of a computer that implements the information processing device, investor terminal, and advisor terminal of the information processing system according to the present embodiment. [Diagram 3] FIG. 2 is a block diagram for explaining an outline of the functions of an information processing device in the information processing system according to the present embodiment. [Figure 4] FIG. 13 is a diagram for explaining an outline of a tag processed by the information processing system according to the present embodiment. [Diagram 5] FIG. 13 is a diagram for explaining an outline of an advisor tag processed in the information processing system according to the present embodiment. [Figure 6] FIG. 13 is a diagram for explaining an outline of investor information processed by the information processing system according to the present embodiment. [Figure 7] FIG. 13 is a diagram for explaining an outline of an investor tag processed by the information processing system according to the present embodiment. [Figure 8] 10 is a flowchart outlining the processing of the information processing system according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Next, an information processing system according to an embodiment of the present invention will be described with reference to FIGS.

[0018] 1 is a block diagram for explaining an outline of the configuration of an information processing system according to an embodiment of the present invention. As shown in the figure, the information processing system 10 mainly comprises an information processing device 20, an investor terminal 30, an advisor terminal 40, and a large-scale language model 50, which are connected to each other so as to be accessible via a network N such as the Internet.

[0019] In this embodiment, the information processing device 20 is managed by an operator 1 that provides a service using the information processing system 10, the investor terminal 30 is owned by an investor 2 that uses a service using the information processing system 10, and the advisor terminal 40 is managed by an advisor 3 that uses a service using the information processing system 10.

[0020] On the other hand, the large-scale language model 50 is managed by an arbitrary business operator that provides a service of providing the large-scale language model 50, and in this embodiment, it is mainly accessed via an information processing device 20 managed by the business operator 1.

[0021] Here, in this embodiment, the service provided by the business operator 1 is a service of proposing to the investor 2 an advisor 3 with specialized knowledge and experience regarding investment through matching based on the request of the investor 2, and the advisor 3 is assumed to be an expert such as an IFA (independent financial advisor) belonging to a financial product intermediary service.

[0022] In this embodiment, the information processing device 20, the investor terminal 30, and the advisor terminal 40 are implemented by computers having substantially the same hardware configuration, such as desktop or notebook computers.

[0023] 2 is a block diagram illustrating an outline of the configuration of a computer. As shown in the figure, the computer includes a processor 101, a memory 102, a storage 103, a transmission / reception unit 104, and an input / output unit 105, which are electrically connected to each other via a bus 106.

[0024] The processor 101 is an arithmetic unit that controls the operation of the computer, controls the transmission and reception of data between each element, and performs processes required for executing application programs.

[0025] In this embodiment, the processor 101 is, for example, a CPU (Central Processing Unit), and executes application programs deployed in a memory 102 (described below) to perform various processes.

[0026] In this embodiment, the memory 102 is implemented as a main storage device configured as a volatile storage device such as a dynamic random access memory (DRAM).

[0027] This memory 102 is used as a working area for the processor 101, and also stores a basic input / output system (BIOS) that is executed when the computer is started, various setting information, and the like.

[0028] The storage 103 stores programs, information used in various processes, and the like.

[0029] The transmission / reception unit 104 connects the computer to the network N, and in this embodiment, the information processing device 20, the investor terminal 30 and the advisor terminal 40 are connected so that they can communicate with each other via the network N via the transmission / reception unit 104.

[0030] This transmitting / receiving unit 104 may be compatible with a wireless communication standard such as Wi-Fi, or may be equipped with a short-range communication interface such as Bluetooth (registered trademark) or BLE (Bluetooth Low Energy).

[0031] The input / output unit 105 is an interface to which input / output devices such as a keyboard, a mouse, or a display are connected.

[0032] The bus 106 transmits, for example, address signals, data signals, and various control signals between the processor 101, memory 102, storage 103, transmission / reception unit 104, and input / output unit 105 that are connected thereto.

[0033] 3 is a block diagram illustrating an outline of functions of the information processing device 20. As illustrated, in this embodiment, the information processing device 20 includes a tag storage unit 20a, an advisor tag storage unit 20b, a question providing unit 20c, an investor information receiving unit 20d, an investor information storage unit 20e, a providing unit 20f, an investor tag receiving unit 20g, an investor tag storage unit 20h, a score assigning unit 20i, a matching unit 20j, and a recommendation unit 20k.

[0034] The tag memory unit 20a, the advisor tag memory unit 20b, the investor information memory unit 20e and the investor tag memory unit 20h are realized by partitioning the memory area of ​​the storage 103, and the question providing unit 20c, the investor information receiving unit 20d, the providing unit 20f, the investor tag receiving unit 20g, the score assigning unit 20i, the matching unit 20j and the recommendation unit 20k are realized by executing programs stored in the memory 102 of the information processing device 20 by the processor 101.

[0035] 4 is a diagram for explaining an outline of tags stored in the tag storage unit 20a. In this embodiment, the tags are text information set in the advisor 3, and as shown in the figure, are composed of terms frequently used in connection with asset management.

[0036] Specifically, the attributes of Advisor 3, such as "Available after 7 p.m. on weekdays," "Available on weekends and holidays," "Suitable for RT Securities (unique name)," "Good at bonds," "Good at stocks," "Good at people in their 60s," "From a major securities company," "Extensive experience (10 years or more)," etc., are composed of terms related to asset management.

[0037] 5 is a diagram for explaining an outline of the advisor tags stored in the advisor tag storage unit 20b. The advisor tags are text information extracted from the tags stored in the tag storage unit 20a and set for each advisor 3 based on the attributes of the advisor 3 regarding asset management.

[0038] Specifically, for example, for advisor 3a, tags based on the attributes of advisor 3a, such as "Available after 7 p.m. on weekdays," "Available on weekends and holidays," "Supports RT Securities (unique name)," "Specializes in people in their 50s," and "Female," are set as advisor tags.

[0039] On the other hand, for advisor 3b, tags based on the attributes of advisor 3b, such as "Available after 7 p.m. on weekdays," "Supports RT Securities (unique name)," "Specializes in people in their 40s," and "Male," are set as advisor tags.

[0040] Furthermore, for advisor 3c, tags based on the attributes of advisor 3c, such as "Available on weekends and holidays", "Specialized in investment trusts", "Specialized in people in their 60s", "Veteran (15 years or more)", etc., are set as advisor tags.

[0041] This advisor tag may be, for example, a tag selected by each advisor 3 from among tags via the advisor terminal 40 and set as the advisor tag, or it may be set automatically based on information regarding the attributes of the advisor 3 managed by the information processing device 20.

[0042] 3 executes a process of providing preset questions regarding investment to the investor 2. In this embodiment, the questions are provided to the investor 2 by transmitting the questions generated as electromagnetic data to the investor terminal 30.

[0043] In this embodiment, this question consists of multiple items such as, for example, "Do you have any concerns about investing?", "Is there anything you would like from an investment advisor?", "How much do you own assets?", and "What is your annual income?".

[0044] In this embodiment, the investor information receiving unit 20d executes a process of receiving, as investor information, replies from the investor 2 to questions provided to the investor 2 (reception process). The investor information received in this reception process is stored in the investor information storage unit 20e.

[0045] 6 is a diagram for explaining an outline of the investor information stored in the investor information storage unit 20e. This investor information is composed of answers to questions from the investor 2, and in this embodiment, is generated from text information input by the investor 2 via the investor terminal 30.

[0046] This investor information is composed based on the investment consultation matters answered in response to questions, requests to the advisor 3, or the attributes of the investor 2, such as the investor's 2 annual income, assets, occupation, age, etc.

[0047] Specifically, as shown in the figure, the investor information of investor 2a is composed of text information including such items as "I have exercised the stock options I received when the company went public...I would prefer to discuss this on the weekend."

[0048] On the other hand, the investor information of investor 2b is composed of text information such as, for example, "37 years old, female, unmarried...I would like to consult with you about my portfolio and future investment destinations...".

[0049] Furthermore, the investor information of investor 2c is composed of text information including such matters as "I would like to manage 200 million yen of surplus funds with reduced risk. I would like to consult with someone with a lot of experience or someone from a private bank."

[0050] In this embodiment, the providing unit 20f shown in FIG. 3 executes a process of providing the investor information received in the reception process and stored in the investor information storage unit 20e to the large-scale language model 50 described later (provision process).

[0051] In this embodiment, the investor tag receiving unit 20g executes a process of receiving the investor tag generated by the large-scale language model 50. In this embodiment, the received investor tag is stored in the investor tag storage unit 20h.

[0052] 7 is a diagram for explaining an outline of the investor tag stored in the investor tag storage unit 20h. The investor tag is generated based on the investor information, and is text information containing the request of the investor 2 to the advisor 3.

[0053] Specifically, as shown in the figure, the investor tag of investor 2a is composed of text information including items such as "Available on weekends and holidays" and "Veteran (15 years or more)", and the investor tag of investor 2b is composed of text information including items such as "Preferring female" and "S Securities compatible".

[0054] On the other hand, the investor tag for investor 2c is composed of text information such as “Available on weekends and holidays”, “From a private bank”, “Veteran (15 years or more)”, and “Available after 7 p.m. on weekdays”.

[0055] In this embodiment, the score assigning unit 20i shown in Figure 3 performs a process of matching an investor tag with an advisor tag and assigning a score to the matched combination of the investor tag and advisor tag depending on the degree of match between the investor tag and the advisor tag (scoring process).

[0056] Specifically, for example, depending on the degree of match, such as the number of matching tags between investor tags and advisor tags or the percentage of matching tags, higher scores are assigned to combinations of investor tags and advisor tags that match more closely.

[0057] In this embodiment, for example, when comparing the investor tag of investor 2c shown in Figure 7 with the advisor tags of advisors 3a to 3c shown in Figure 5, a score of "4" is assigned depending on the degree of match between the investor tag of investor 2c and the advisor tag of advisor 3a, a score of "3" is assigned depending on the degree of match between the investor tag of investor 2c and the advisor tag of advisor 3b, and a score of "5" is assigned depending on the degree of match between the investor tag of investor 2c and the advisor tag of advisor 3c.

[0058] In this scoring process, for example, a weight may be assigned to any tag, and when a weighted tag matches, a score according to the assigned weight may be assigned.

[0059] In this embodiment, the matching unit 20j shown in Figure 3 executes a process (matching process) of matching investors 2 and advisors 3 based on the scores assigned in the scoring process, and executes a process such as matching combinations whose assigned scores exceed an arbitrary threshold value (e.g., "4").

[0060] In this embodiment, for example, a score of "4" is assigned to the combination of the investor tag of investor 2c and the advisor tag of advisor 3a, a score of "5" is assigned to the combination of the investor tag of investor 2c and the advisor tag of advisor 3c, while a score of "3" is assigned to the combination of the investor tag of investor 2c and the advisor tag of advisor 3b.

[0061] Therefore, in the matching process, a process is executed to match the investor 2c with the advisors 3a and 3c whose investor ratings exceed a preset threshold (for example, "4").

[0062] The recommendation unit 20k executes a process of proposing to the investor 2 an advisor 3 that is matched with the investor 2 in the matching process, and in this embodiment, creates recommendation information and transmits the created recommendation information to the investor terminal 30, thereby proposing the advisor 3 to the investor 2.

[0063] The recommendation information may consist of, for example, the names of the advisors 3 arranged in order of highest score, attributes of the advisors 3 such as their ages, areas of expertise, and speed of response, comments analyzing the needs of the investor 2, or comments regarding the reasons for recommending a particular advisor 3 to the investor 2.

[0064] The large-scale language model 50 shown in FIG. 1 is a model that performs natural language processing using deep learning, and in this embodiment, performs a process of generating investor tags based on investor information transmitted from the information processing device 20.

[0065] Next, an overview of the processing of the information processing system 10 according to the present embodiment will be described.

[0066] 8 is a flowchart for explaining an outline of the processing of the information processing system 10 according to this embodiment. As shown in the figure, first, in step S1, a question regarding investment is transmitted to the investor terminal 30, thereby providing the question to the investor 2.

[0067] In response to this, the investor 2 answers the question and transmits the answer to the information processing device 20. In step S2, the information processing device 20 accepts the answer as investor information (reception process). The investor information accepted in the reception process is stored in the investor information storage unit 20e.

[0068] When the investor information is received, in step S3, the received investor information is provided to the large-scale language model 50 (providing process). The large-scale language model 50 generates investor tags based on the provided investor information.

[0069] Next, in step S4, the information processing device 20 executes a process of accepting the investor tag generated by the large-scale language model 50. The accepted investor tag is stored in the investor tag storage unit 20h.

[0070] In step S5, the investor tag and the advisor tag are collated, and a process of assigning a score to the matched combination of the investor tag and the advisor tag is executed (scoring process). In this embodiment, a score is assigned according to the degree of match between the investor tag and the advisor tag.

[0071] Next, in step S6, the investor 2 and the advisor 3 are matched based on the scores assigned in the scoring process (matching process), and in step S7, recommendation information is created based on the matching process and sent to the investor terminal 30, thereby proposing to the investor 2 an advisor 3 that is a match for the investor 2.

[0072] In this way, a score is assigned by comparing the investor tag relating to the requests of investor 2 to advisor 3 with the advisor tag relating to the attributes of advisor 3. Therefore, for example, by performing a matching process based on the assigned score, good matching between investor 2 and advisor 3 can be achieved.

[0073] In particular, in this embodiment, since investor tags are generated by the large-scale language model 50 based on investor information provided by investors in response to questions, it is expected that appropriate investor tags will be generated even for investors 2 who are unable to adequately verbalize their requests to advisor 3.

[0074] Therefore, since Investor 2 can receive investment advice from Advisor 3 with whom he / she has a good compatibility, it is expected that a favorable investment environment will be created for Investor 2.

[0075] The present invention is not limited to the above-described embodiment, and various modifications are possible without departing from the spirit and scope of the invention.

[0076] In the above embodiment, a case was described in which investor information is generated by text information input to investor 2 via investor terminal 30, but it may also be generated by voice information input to investor 2 via investor terminal 30.

[0077] In the above embodiment, a case has been described in which investor information is provided to the large-scale language model 50 via the information processing device 20, but the investor information may also be configured to be provided directly to the large-scale language model 50 via the investor terminal 30. [Explanation of symbols]

[0078] 1 Operator 2 Investor 3 Advisor 10. Information Processing Systems 20 Information processing device 30 Investor Terminal 40 Advisor Terminals 50 Large-scale language models (language models)

Claims

1. A reception process for receiving answers to preset questions regarding investments from the investors as investor information; a providing process for providing the investor information received in the receiving process to a language model; a scoring process for matching an investor tag related to a request of the investor for an advisor who provides the investment advice to the investor, the investor tag being generated by the language model based on the investor information provided in the providing process, with an advisor tag set based on the attributes of the advisor, and assigning a score; An information processing system that executes the above.

2. executing a matching process for matching the investor with the advisor based on the score assigned in the scoring process; The information processing system according to claim 1 .

3. The scoring process includes: assigning the score based on the degree of match between the investor tag and the advisor tag; 3. The information processing system according to claim 1 or 2.

4. The investor information is: Consists of audio or text information, 3. The information processing system according to claim 1 or 2.

5. An information processing device including a processor and a memory in which a program is stored, When the program is executed by the processor, A reception process for receiving answers to preset questions regarding investments from the investors as investor information; a providing process for providing the investor information received in the receiving process to a language model; a scoring process for matching an investor tag related to a request of the investor for an advisor who provides the investment advice to the investor, the investor tag being generated by the language model based on the investor information provided in the providing process, with an advisor tag set based on the attributes of the advisor, and assigning a score; An information processing device that executes the above.

6. An information processing device implemented by a computer, A reception process for receiving answers to preset questions regarding investments from the investors as investor information; a providing process for providing the investor information received in the receiving process to a language model; a scoring process for matching an investor tag related to a request of the investor for an advisor who provides the investment advice to the investor, the investor tag being generated by the language model based on the investor information provided in the providing process, with an advisor tag set based on the attributes of the advisor, and assigning a score; A program that executes the following.

7. An information processing device implemented by a computer, A reception process for receiving answers to preset questions regarding investments from the investors as investor information; a providing process for providing the investor information received in the receiving process to a language model; a scoring process for matching an investor tag related to a request of the investor for an advisor who provides the investment advice to the investor, the investor tag being generated by the language model based on the investor information provided in the providing process, with an advisor tag set based on the attributes of the advisor, and assigning a score; An information processing method for performing the above.

Images