Information Processing System and Information Processing Method
The information processing system addresses the challenge of determining appropriate business counterparts by analyzing public information and generating recommendations, thereby reducing user labor and improving decision-making accuracy.
Patent Information
- Application Number
- JP2024108684
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2025-06-23
- Estimated Expiration
- 2044-07-05
AI Technical Summary
Companies face significant labor and resource challenges in determining whether to conduct business with potential counterparts, as they need to collect and analyze extensive information to assess appropriateness.
An information processing system that acquires specific target person information from public sources and analyzes it using predefined criteria to generate analysis result information, indicating the degree to which it is recommended to conduct business with a specific target party.
This system automates the process of evaluating potential business partners, reducing the user's labor and enabling accurate decisions based on generated analysis result information without the need for manual data collection and analysis.
Smart Images

Figure 0007696584000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an information processing system for supporting sales and an information processing method using the information processing system.
Background Art
[0002] Conventionally, in a company that conducts business activities in exchange for consideration from customers, it is widely practiced to conduct sales to potential customers. Although the entity conducting sales is not limited to companies, for the sake of simplicity of explanation, the entity conducting sales is assumed to be a company. Regarding this type of sales, various technologies for supporting sales have been proposed. For example, Patent Document 1 describes the following technology. That is, the sales support device 40 calculates the probability that a case with a predetermined attribute that has reached a predetermined phase will result in a contract thereafter, based on case history information including the attributes of the case and the history of the transition of the phase of progress from the start of business with the business partner to the conclusion or non - conclusion of the case. The above technology is described. According to the technology of Patent Document 1, since the future probability of concluding a contract for ongoing cases can be estimated, it becomes possible to support sales by the user.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] Regarding business operations by a company, there are appropriate and inappropriate parties as counterparts for conducting business. As an example, for a company operating a news website that conducts business regarding the submission of advertisements to its news website, a party engaged in business activities with high affinity to the news website is appropriate, while others are not. For example, if the user group of the news website is in their 20s to 40s, and a certain party is engaged in selling toys for infants, the affinity between the news website and the business activities of that party is low. In this case, it can be said that that party is not an appropriate counterpart for conducting business. As another example, for a company conducting a service as a business and looking for counterparts for promoting the use of its service, a company that can benefit from using the service (i.e., a company with the potential to use the service) is appropriate, while a company that cannot benefit at all from using the service (i.e., a company with no potential to use the service) is not appropriate.
[0005] Based on the above, a company conducting business has a need to appropriately determine whether or not to conduct business with a party who can be a counterpart for business (hereinafter referred to as "target party") instead of approaching everyone indiscriminately. To meet this need, the following operations have been conventionally carried out. That is, the person in charge collects information on the target party from the target party's website, sources where information on each target party is published, reports by external research institutions, and other information sources, analyzes the collected information, and makes a determination on whether or not to conduct business. However, in the case of this operation, since the person in charge needs to check the information and recognize the appropriateness of the target party as a business counterpart based on the information before making a determination on whether or not to conduct business, the labor (effort, time, cost, etc.) required for such determination is significant.
[0006] The present invention has been made to solve such problems, and aims to enable a user to accurately determine whether or not to conduct business with a certain target party while reducing the user's labor.
Means for Solving the Problem
[0007] In order to solve the above problems, the present invention acquires specific target person information based on specific public information regarding a specific target person among the public information regarding the disclosed target persons, and is recommended to conduct business with the target person regarding a predetermined matter. Analysis result information indicating the degree to which it is recommended to conduct business with a specific target person is acquired based on the analysis information and the specific target person information to be referred to when analyzing the content of the public information from the viewpoint of whether or not.
Effect of the Invention
[0008] According to the present invention configured as described above, based on the specific target person information based on the specific public information disclosed regarding the specific target person, and "the analysis information to be referred to when analyzing the content of the public information from the viewpoint of whether it is recommended to conduct business with the target person regarding a predetermined matter", "Analysis result information indicating the degree to which it is recommended to conduct business with a specific target person" is automatically generated. Therefore, the user can recognize how recommended it is to conduct business with a specific target person by referring to the analysis result information without collecting the information of the specific target person by himself / herself or analyzing the collected information, and can accurately determine whether or not to conduct business with the specific target person based on this recognition. That is, according to the present invention, it is possible to enable the user to accurately determine whether or not to conduct business with a certain target person while reducing the labor of the user.
Brief Description of the Drawings
[0009]
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Mode for Carrying Out the Invention
[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a control system 1 (information processing system) according to the present embodiment. As shown in FIG. 1, the control system 1 includes an information processing server 2 (information processing system, server, computer) and a user terminal 3 (information processing system, terminal, computer). Both the information processing server 2 and the user terminal 3 can be connected to a network N including the Internet, a telephone network, and other communication networks. A large language model server 4 is connected to the network N.
[0011] The information processing server 2 is a server device connected to the network N. The information processing server 2 has a function of providing various services to be described later to the user. Hereinafter, the services provided by the information processing server 2 are referred to as "the present services". The user has registered for the present services and can log in to the account of the present services using his / her own login information (for example, a combination of a login ID and a password). In FIGS. 1 and 2, the information processing server 2 is represented by one block. However, this does not mean that the information processing server 2 is composed of a single server device. For example, the information processing server 2 may be composed of a plurality of server devices. In this case, the server devices constituting the information processing server 2 may include a web server or a web application server. Also, in this case, the information processing server 2 may be configured to include a plurality of server devices that are load-balanced by a load balancer. In this case, each of the server information processing units 10 of the plurality of server devices, or a plurality of combinations, functions as an "information processing unit".
[0012] The user terminal 3 is a terminal used by the user. In the present embodiment, a person who can receive the present services provided by the information processing server 2 is referred to as a "user". The type of the user terminal 3 may be anything. For example, a desktop computer, a notebook computer, or a tablet computer (including a so-called smartphone) can function as the user terminal 3.
[0013] The large language model server 4 is a server equipped with a large language model 5 (language model). The provider, type, and machine learning method of the large language model 5 are not limited. The large language model server 4 has a function of receiving a prompt (question text) from an external device, generating an answer text for the prompt by the function of the large language model 5, and responding to the external device with the answer text. The information processing server 2 is appropriately authorized to use the large language model server 4.
[0014] FIG. 2 is a block diagram showing an example of the functional configuration of the information processing server 2 and the user terminal 3. As shown in FIG. 2, the information processing server 2 includes, as functional components, a server information processing unit 10 (information processing unit), a server communication unit 11, and a server storage unit 12. The user terminal 3 includes, as functional components, a terminal information processing unit 13 (information processing unit), a terminal communication unit 14, a terminal display unit 15, a terminal input unit 16, and a terminal storage unit 17.
[0015] The server information processing unit 10 includes a processing device including a processor and a primary storage device. The server information processing unit 10 executes processing by reading a program stored in the storage area (which may be another storage area) of the server storage unit 12 into the primary storage device and executing it by the processing device. That is, the server information processing unit 10 executes processing by the cooperation of hardware and software. The server communication unit 11 includes a communication device including a communication control device and a network interface. The server communication unit 11 communicates with an external device by the communication device under the control of the server information processing unit 10. Hereinafter, it is assumed that the communication by the information processing server 2 is appropriately executed by the server communication unit 11, and the description of the communication is omitted. The server storage unit 12 stores data in a non-volatile memory. The non-volatile memory is, for example, a hard disk drive (which may be another magnetic storage device), a ROM, or a flash memory.
[0016] The terminal information processing unit 13 includes a processing device including a processor and a primary storage device. The terminal information processing unit 13 executes processing by reading a program stored in a storage area (which may be another storage area) of the terminal storage unit 17 into the primary storage device and executing it. That is, the terminal information processing unit 13 executes processing through the cooperation of hardware and software. The terminal communication unit 14 includes a communication device including a communication control device and a network interface. The terminal communication unit 14 communicates with an external device by the communication device under the control of the terminal information processing unit 13. Hereinafter, it is assumed that the communication by the user terminal 3 is appropriately executed by the terminal communication unit 14, and the description of the communication is omitted. The terminal display unit 15 includes a liquid crystal panel, an organic EL panel, and other display devices. The terminal display unit 15 displays an image on the display device under the control of the terminal information processing unit 13. The terminal input unit 16 detects an input to the input device and outputs the detection result to the terminal information processing unit 13. The input device is, for example, a keyboard, a mouse, a touch panel, or a camera. The terminal storage unit 17 stores data in a non-volatile memory.
[0017] Next, the operation of the control system 1 will be described by dividing it into scenes. In the following description, a "news site operator" is assumed as one of the users. A news site operator is a company that operates a specific news site, which is one of the news sites. Hereinafter, for convenience of explanation, a person belonging to the news site operator is also referred to as a "news site operator", and is expressed, for example, as "the news site operator uses the user terminal 3". The specific news site is a site composed of one or more news pages 20 on which news articles are posted. The provision of the specific news site is performed via the client's browser or a dedicated application.
[0018] FIG. 3 is a diagram showing an example of the news page 20. As shown in FIG. 3, on the news page 20, news blocks 21 in which the titles of news articles, related photos, etc. are displayed are arranged and displayed. Each news block 21 is selectable, and when one news block 21 is selected, the details of the news article corresponding to the one news block 21 are displayed. Also, on the news page 20, an advertisement block 22 may be displayed between the news blocks 21. The advertisement block 22 is a block in which information related to advertisements is displayed, and is displayed based on an advertisement published by an advertiser. The advertisement block 22 is selectable, and when the advertisement block 22 is selected, detailed information about the advertisement is displayed on the screen, or the screen of the corresponding link destination is displayed. However, the method of providing advertisements on the news page 20 is not limited to the exemplified method. For example, an advertisement banner may be displayed at a predetermined position on the page.
[0019] News site operators receive advertisements from advertisers and display advertisement blocks 22 corresponding to the advertisements on news pages 20 to obtain compensation from the advertisers. Therefore, news site operators have a need to conduct sales activities for potential customers who can become advertisers and conclude contracts for the submission of advertisements. On the other hand, among the parties with whom news site operators conduct sales activities, there are appropriate and inappropriate parties. More specifically, a specific news site has characteristics in terms of user demographics, target demographics, service delivery methods, the atmosphere of news page 20, and other matters. The specific news site according to this embodiment has the characteristic that the user demographic is business people in their 20s to 40s, and has the characteristic that it targets people interested in business / finance / economics and technology in terms of the target demographic. Furthermore, news site operators have expectations regarding the content / direction of advertisements provided via the news site, the form of advertisers (such as scale, capital, or location), and other matters. Based on this, those who highly match the characteristics of the specific news site and the expectations of the news site operator are highly suitable as parties with whom the news site operator conducts sales activities, and those with a low degree of match are less suitable. Based on the above, news site operators have a need to accurately determine whether to conduct sales activities for potential customers (hereinafter referred to as "target parties") who can become parties with whom they conduct sales activities, rather than conducting sales activities indiscriminately for anyone. As will become clear later, the information processing server 2 according to this embodiment has a function to support the above determination of the news site operator.
[0020] <Scene S1 where the news site operator registers analysis information> First, the operation of the control system 1 in the scene S1 where the news site operator registers analysis information will be described. The news site operator launches the browser of the user terminal 3 and instructs access to a predetermined URL of the information processing server 2. In response to the instruction, the terminal information processing unit 13 accesses the predetermined URL using the function of the browser. When there is an access to the predetermined URL, the server information processing unit 10 of the information processing server 2 appropriately authenticates in a predetermined method and then responds to the terminal information processing unit 13 with display data (including an HTML file and data associated with the HTML file) for displaying the main screen. The terminal information processing unit 13 receives the display data and displays the main screen on the terminal display unit 15 based on the display data using the function of the browser.
[0021] Hereinafter, the detection of operations on various screens displayed on the terminal display unit 15, the acquisition of information input to various screens, the dynamic change of various screens, and various other processes related to the screens are appropriately executed by the server information processing unit 10 and the terminal information processing unit 13 using CGI, JavaScript (registered trademark), Ajax, DHTML, and other existing technologies, and no detailed description will be given. Also, the fact that the server information processing unit 10 transmits display data to the terminal information processing unit 13 and causes the terminal display unit 15 to display a screen / user interface based on the display data may be expressed as "the server information processing unit 10 causes the terminal information processing unit 13 to display a ○○ screen / ○○ user interface", or simply "the server information processing unit 10 displays a ○○ screen / ○○ user interface".
[0022] The news site operator performs a predetermined operation on the main screen displayed on the terminal display unit 15 to instruct the display of the analysis information input screen 23. The server information processing unit 10 detects the instruction and causes the terminal information processing unit 13 to display the screen. FIG. 4 is a diagram showing an example of the analysis information input screen 23. The analysis information input screen 23 is a screen (user interface) into which analysis information can be input. In this way, the terminal information processing unit 13 has a function of providing a user interface into which analysis information can be input. As will be made clear later, the server information processing unit 10 uses a press release published on the Internet to obtain information regarding the appropriateness of the news site operator conducting business with the company that is the source of the press release. And the analysis information is information to be referred to when analyzing the content of a press release (public information) from the perspective of whether it is recommended for the "company that is the source of the press release" (the target) to conduct business regarding the publication of an advertisement (predetermined matters) on a specific news site.
[0023] In this embodiment, the analysis information includes positive element information indicating elements that are factors for positively judging about conducting business with the company that is the source of the press release, and negative element information indicating elements that are factors for negatively judging. On the analysis information input screen 23, an input area 24 for inputting positive element information and an input area 25 for inputting negative element information are respectively provided. In the input area 24, one or more pieces of positive element information can be input as a text for each group of content (however, it is not necessary to input any). The same applies to the input area 25. It is assumed that the news site operator sets the analysis information in reflection of the characteristics of the specific news site and the wishes of the news site operator regarding the advertisement / advertiser.
[0024] Positive element information is as follows, for example. The content of the press release is an introduction of IT-related products or services / The content of the press release is an introduction of new products or services / The content of the press release is an introduction of products or services for BtoB / The company that issued the press release is listed or has a capital of over 50 million yen / The content of the press release is an introduction of a service that utilizes ○○ of conversational AI. Negative element information is as follows, for example. The content of the press release is an introduction of limited-time products / The content of the press release is an introduction of campaigns, events or fairs / The content of the press release is an introduction of the opening of stores or facilities / The content of the press release is an introduction related to food and beverage / The capital is 50 million yen or less / There is no description of the capital.
[0025] When the news site operator inputs positive element information and negative element information into input area 24 and input area 25 and then selects the confirmation button 26 when finalizing the input. When the confirmation button 26 is selected, the terminal information processing unit 13 transmits the input analysis information to the server information processing unit 10 as data in a predetermined format. When the server information processing unit 10 receives and acquires the analysis information, it accesses the user management database 27 in the server storage unit 12. The user management database 27 is a database in which records are registered for each user. (A) in FIG. 5 shows the content of one record in the user management database 27. In the following description, <f>is a symbol that conveniently indicates that the corresponding term is an item / field in which a value is stored. As shown in (A) of FIG. 5, one record of the user management database 27 has, as fields, a user ID <f>And, user management <f>and analysis information <f>comprising. User ID <f>It stores a user ID for uniquely identifying each user. User management <f>It stores the name, contact information, and other management information of the corresponding user. Information for analysis <f>Analysis information is stored therein.
[0026] After accessing the user management database 27, the server information processing unit 10 analyzes the information for the record corresponding to the news site operator. <f>Store the received analysis information. Thus, the registration of the analysis information by the news site operator is completed. Although details are omitted, the news site operator can freely update the content of the analysis information registered in the user management database 27 at any time using the user interface provided by the terminal information processing unit 13 under the control of the server information processing unit 10.
[0027] <Scene S2 where analysis result information for the news site operator is obtained based on a press release> Next, the operation of the control system 1 in scene S2 where analysis result information for the news site operator is obtained based on a press release will be described. The flowchart FA in FIG. 6 shows an operation example of the information processing server 2 in scene S2.
[0028] As shown in the flowchart FA, the server information processing unit 10 monitors whether a new press release page 30 has been published on a specific press release site (step SA1). Hereinafter, the process of step SA1 will be described in detail. In this embodiment, there is a specific press release site on the Internet as one of the press release sites. The specific press release site is a site that publishes a press release page 30 on which press releases issued by companies are posted at any time. A press release generally means an official document issued by a company or organization, but in this embodiment, it is not limited to this, and widely includes information published through the specific press release site. A press release generally includes an announcement or promotion by the issuing company. Note that the entity that issues a press release is not limited to a "company", but for simplicity of explanation, the entity that issues a press release is assumed to be a company. Hereinafter, the company that issues a press release is referred to as the "issuing company". The issuing company corresponds to the "target person". Also, a press release corresponds to "public information about the target person (published on the Internet)".
[0029] FIG. 7 is an example of a press release page 30. As shown in FIG. 7, the press release page 30 includes a press release title (reference numeral 31), the name of the issuing company (reference numeral 32), the release date and time (reference numeral 33), the body of the press release (reference numeral 34), and information indicating basic matters regarding the issuing company (reference numeral 35). Hereinafter, the press release title is referred to as the "press release title", and the body of the press release is referred to as the "press release body". The press release body is the part where the specific content of the press release is described, and may include photographs, illustrations, link information, and other accompanying information in addition to the information in the text. Also, the basic matters regarding the issuing company are referred to as "company basic information".
[0030] On a specific press release site, the company basic information is composed of item values set for a plurality of predetermined items (hereinafter referred to as "company basic items"). The issuing company can also leave the item value blank for any item. The company basic items include the company name <f>, URL <f>, Industry type <f>, phone number <f>, Representative's Name <f>, listing <f>, Capital <f>, and establishment <f>includes. URL <f>The item value is the URL of the company's homepage. Also, for listed companies <f>The item value indicates information about the market if the company is listed, and indicates that it is not listed if the company is not listed.
[0031] When a new press release page 30 is published on a specific press release site, update information including the URL of the page, the press release title, a summary of the press release (a summary different from the press release summary described later), the publication date and time, and other elements is notified to the server information processing unit 10. This notification is performed by RSS, and the settings necessary to realize this notification are appropriately made in advance. Then, in step SA1, the server information processing unit 10 monitors whether the notification has been received, and if received, determines that a new press release page 30 has been published.
[0032] If it is determined in step SA1 that a new press release page 30 has been published (step SA1: YES), the server information processing unit 10 executes a press release acquisition process (step SA2). Specifically, the server information processing unit 10 accesses the URL of the notified update information, and acquires the HTML data (including data attached thereto; the same applies hereinafter) of the press release page 30 at the access destination as "press release data". That is, the HTML data that is the basis of the newly published press release page 30 is the "press release data". Each piece of information on the press release page 30 is structured according to HTML and described in the press release data. In particular, for the basic company information, each item name of the basic company items is defined using "description list tags" (dl, dt, dd tags), and the item values of each item are described.
[0033] The press release corresponding to the press release data acquired in step SA2 corresponds to "specific public information", and the company that issued the press release corresponds to the "specific target person". Hereinafter, the company that issued the press release corresponding to the press release data acquired in step SA2 is particularly referred to as the "specific issuing company". The processes in step SA3 and subsequent steps are processes executed for the specific issuing company.
[0034] After step SA2, the server information processing unit 10 executes extraction processing (step SA3). Specifically, the server information processing unit 10 extracts and obtains the press release title and the press release text from the press release data. Note that the press release text is extracted as "information by text" from which elements other than text (for example, still images or videos) are removed. In the present embodiment, the server information processing unit 10 executes the processing of step SA3 by the function of a program that causes a computer to analyze HTML data and extract desired information. However, the server information processing unit 10 may be configured to execute the processing of step SA3 using the function of the large language model 5. In the case of this configuration, the server information processing unit 10 outputs a prompt requesting the large language model 5 to extract the press release title and the press release text from the press release data, and obtains the press release title and the press release text by obtaining the response sentence.
[0035] After step SA3, the server information processing unit 10 executes summary acquisition processing (step SA4). The summary acquisition processing is processing for obtaining a summary of the press release corresponding to the press release (hereinafter referred to as "press release summary"). Hereinafter, the processing of step SA4 will be described in detail. The server information processing unit 10 generates and obtains a summary prompt (second prompt), which is a prompt for the large language model 5. The summary prompt includes the press release title and the press release text, and is a prompt for requesting to generate a summary thereof. The summary prompt includes the following requests. · Generate a summary based on the press release title and the press release text. · Make the number of characters of the press release summary about 800 characters. · Ensure that proper nouns existing in the press release title and the press release text are included in the press release summary.
[0036] The reason for including proper nouns present in the press release title and the press release body in the press release summary will become clear later. The press release title and the press release body correspond to "related articles" including announcements or promotions by specific targets. Hereinafter, an example of a summary prompt is shown. However, the following example is for the purpose of showing the outline of the summary prompt, and the validity as a prompt is ignored. This also applies to other examples of prompts.
[0037] ***Start of the summary prompt*** You are an expert in summarizing press releases. On a press release site, a press release with the following <title> and the following <body> has been published. Please summarize this press release in about 800 words. However, please make sure that the summary includes the proper nouns present in the <title> and the <body>. <Title> Made by a new manufacturing method... <Body> Always with a new taste... ***End of the summary prompt***
[0038] The server information processing unit 10 according to this embodiment generates and acquires a summarization prompt by the function of the program stored in the server storage unit 12. This program has a function of generating a summarization prompt by embedding a press release comment and a press release text into the format of a pre-prepared summarization prompt. However, the method by which the server information processing unit 10 acquires the summarization prompt is not limited to the method exemplified in this embodiment. For example, the server information processing unit 10 may be configured to request the large language model 5 to generate a summarization prompt and acquire the summarization prompt as a response sentence to the request. Also, for example, the following configuration may be used. That is, the summarization prompt is generated by an external device or by artificial means and stored in the server storage unit 12 (or another storage unit accessible by the server information processing unit 10). The server information processing unit 10 accesses the server storage unit 12 and acquires the summarization prompt. The above configuration may be used. The above points regarding the acquisition of the summarization prompt are the same for the acquisition of other prompts.
[0039] Also, the content of the exemplified summarization prompt is merely an example. In particular, in the above example, the press release title and the press release text extracted from the press release are included in the summarization prompt, but instead of or together with these, all or part of the press release data may be included in the summarization prompt while maintaining the state of HTML data. The above points regarding the content of the summarization prompt are the same for other prompts.
[0040] After obtaining the summary prompt, the server information processing unit 10 transmits the summary prompt to the large language model server 4. Note that the process of transmitting a prompt (not limited to the summary prompt) to the large language model server 4 corresponds to the process of providing the prompt to the large language model 5. The large language model server 4 receives the summary prompt and generates a press release summary (answer text) for the summary prompt by the function of the large language model 5. The large language model server 4 transmits the press release summary to the server information processing unit 10. The server information processing unit 10 receives and obtains the press release summary. As described above, the server information processing unit 10 generates the press release summary using the function of the large language model 5. The above is the detailed process of step SA4.
[0041] After step SA4, the server information processing unit 10 executes specific basic information acquisition processing (step SA5). The specific basic information acquisition processing is a process of acquiring specific basic information. The specific basic information is information in which an item name and an item value are associated with each of the corporate basic items of a specific issuing company. For example, when the corporate basic information of a specific issuing company is as illustrated in FIG. 7, the specific basic information is "{Company name: ○○ Co., Ltd.}, {URL: https: / / ...}, {Industry: Manufacturing},...". In step SA5, the server information processing unit 10 analyzes the press release data by the function of a predetermined program, generates and acquires the specific basic information. However, the server information processing unit 10 may be configured to acquire the specific basic information using the function of the large language model 5. Also, the server information processing unit 10 may be configured to acquire the specific basic information generated externally.
[0042] After step SA5, the server information processing unit 10 executes analysis processing (step SA6). The analysis processing is a process of obtaining analysis result information indicating the degree to which it is recommended to conduct business with a specific issuing company, and evaluation reason information indicating the reason why the analysis result information has that value. Hereinafter, the processing of step SA6 will be described in detail. First, the server information processing unit 10 accesses the user management database 27 and obtains analysis information corresponding to the news site operator. In this way, the server information processing unit 10 has a function of obtaining analysis information based on the information input on the analysis information input screen 23 (user interface). Next, the server information processing unit 10 generates and obtains an analysis prompt (first prompt), which is a prompt for the large language model 5. The analysis prompt includes analysis information and specific target person information, and is a prompt that requests to analyze the specific target person information by reflecting the analysis information to generate analysis result information. In the present embodiment, the specific target person information is a combination of the press release title obtained in step SA3, the press release summary obtained in step SA4, and the specific basic information obtained in step SA5. The analysis prompt includes the following requests. · Analyze the specific target person information from the perspective of whether it is recommended to conduct business with the specific issuing company regarding the publication of advertisements on a specific news site. · Reflect the analysis information in the above analysis. · As a result of the above analysis, generate analysis result information indicating the degree of recommendation to conduct business on a five-level evaluation, and evaluation reason information indicating the reason why the analysis result information has that value. In addition, the analysis prompt includes, as information on the business side, the news site operator and information regarding the specific news site. Hereinafter, an example of the analysis prompt will be shown.
[0043] ***Start of the analysis prompt*** # Prerequisites · You are a person in charge of business regarding the publication of advertisements on a news site. · The news site targets business people in their 20s to 40s as its user base and those interested in business / finance / economy and technology. The news site is widely popular in Japan. · You are looking for a partner to conduct sales regarding the publication of advertisements on the news site using press releases. Based on the following press release, please determine whether sales should be conducted with the company that issued the press release regarding the publication of advertisements on the news site. # Definition of variables and goal setting 1. Content of the press release · Title: Selling ○○ Beer Made by a New Manufacturing Method · Article text: Always challenging new tastes... 2. Basic information of the company that issued the press release · Company name: ××× Co., Ltd. · URL: https… · Industry: Manufacturing … 3. Criteria for judging whether to conduct sales ● Factors that positively influence the decision to conduct sales · The content of the press release is an introduction of IT-related products or services. · The content of the press release is an introduction of new products or services. · The content of the press release is an introduction of products or services for BtoB. · The company that issued the press release is listed or has a capital of over 50 million yen. · The content of the press release is an introduction of a service using ○○, which is the conversational AI service of ○○ company. ● Factors that negatively influence the decision to conduct sales · The content of the press release is an introduction of limited-time products. · The content of the press release is an introduction of campaigns, events or fairs. · The content of the press release is an introduction of the opening of stores or facilities. · The content of the press release is related to food and beverage introductions. · The capital is 50 million yen or less. · There is no description of the capital. 4. Please return the judgment result (5: Should be proposed, 4: It is better to propose, 3: Neither can be said, 2: It is better not to propose, 1: Should not be proposed) and the reason for the judgment in JSON format. · Please make a judgment reflecting all of the elements that are positive factors for judging to conduct business and the elements that are negative factors for judging to conduct business. · The judgment result should be made in a five - level evaluation of "5: Should be proposed, 4: It is better to propose, 3: Neither can be said, 2: It is better not to propose, 1: Should not be proposed". · Please include important keywords and information in the reason for the judgment. # Execution process of the procedure 1. Read the given "1. Content of the press release" and "2. Basic information of the company that is the source of the press release". 2. Based on the "3. Judgment guidelines for determining whether to propose", determine whether to conduct business regarding the publication of advertisements on news sites. 3. Output the result. # Regarding the output of the result · The result should be in JSON format. · JSON format { ”score”:”Judgment result (numerical value from 1 to 5)”, ”reason”:”Reason for the judgment” } *** End of the analysis prompt ***
[0044] In an example of the analysis prompt shown above, the press release title is stored in "·Title", and the press release summary is stored in "·Article Body". "2. Basic information of the company that issued the press release" is described based on specific basic information. "3. Criteria for determining whether to conduct business" is described based on analysis information.
[0045] After obtaining the analysis prompt, the server information processing unit 10 transmits the analysis prompt to the large language model server 4. The large language model server 4 receives the analysis prompt and generates analysis result information and evaluation reason information for the analysis prompt by the function of the large language model 5. The analysis result information corresponds to the "Judgment Result" of the exemplified analysis prompt. That is, the analysis result information is information indicating the degree of recommending business on a five-level evaluation. The evaluation reason information corresponds to the "Reason for Judgment" of the exemplified analysis prompt. That is, the evaluation reason information is information indicating the reason why the analysis result information has that value. The large language model server 4 transmits the analysis result information and the evaluation reason information to the server information processing unit 10. The server information processing unit 10 receives and acquires these information. As described above, the server information processing unit 10 generates the analysis result information and the evaluation reason information by using the function of the large language model 5. The above is the detail of step SA6.
[0046] The analysis result information is information indicating the degree to which it is recommended to conduct business with a specific issuing company. And in this embodiment, since the analysis result information is obtained by the above method, there are the following advantages. That is, the press release is analyzed as the analysis target, and the analysis result information is derived. Since the press release includes announcements or promotions by a specific issuing company, it is assumed that it contains many useful elements for determining whether the specific issuing company is appropriate as a partner for a news site operator to conduct business. From this point, the accuracy of the analysis result information can be increased. Also, instead of the press release text, the press release summary is analyzed as the analysis target, and the analysis result information is derived. Here, the press release text may include redundant parts or parts of low importance in determining the appropriateness of business due to the nature of the article, but the press release summary is a state in which the gist and key points of the text are summarized. By analyzing such a press release summary as the analysis target, the accuracy of the analysis result information can be increased. Also, the press release summary is made to include proper nouns existing in the press release title and the press release text. Here, the press release includes announcements or promotions by the issuing company. And in announcements or promotions, often the proper nouns included therein become important points / terms that represent or symbolize the announcements or promotions. Based on this, by including the proper nouns existing in the press release title and the press release text in the press release summary, the accuracy of the analysis result information can be increased. Also, the specific basic information is also included in the analysis target. For this reason, it is possible to set analysis information taking into account the specific basic information, and the wishes of the news site operator can be reflected in the analysis result information more flexibly.
[0047] After step SA6, the server information processing unit 10 executes email text acquisition processing (step SA7). Hereinafter, the processing of step SA7 will be described in detail. The server information processing unit 10 generates and acquires an email text prompt (third prompt), which is a prompt for the large language model 5. The email text prompt includes specific target person information and requests to generate the text of a business email that can be used when a news site operator conducts business with a specific publishing company reflecting the information. As described above, in this embodiment, the specific target person information includes a press release title, a press release summary, and specific basic information. The email text prompt includes the following requests. · Generate the text of a business email that can be used when conducting business regarding the submission of advertisements to a specific news site to the company that is the source of the press release, reflecting the specific target person information. The text of the business email corresponds to "business text that can be used when a user conducts business with the target person". Hereinafter, an example of the email text prompt will be shown.
[0048] ***Start of the email text prompt*** · You belong to the company indicated by the following <basic information of your company> and are in charge of conducting business regarding the submission of advertisements to the news site. · The news site targets business people in their 20s to 40s as its user group and those interested in business / finance / economy and technology. The news site is a widely popular site in Japan. · You have read the following <press release> and are trying to create a business email for conducting business regarding the submission of advertisements to the news site to the company that is the source of the <press release>. The basic information of the said source company is as indicated by <basic information of the source company>. Please consider the text of the business email to be sent to the said source company based on the <press release>. <Press release> · Title: Made by the new manufacturing method... · Article text: Always with a new taste... <Basic information of the source company> Company Name: ○○ Co., Ltd. … <Your Company's Basic Information> Company Name: △△ … ***End of Email Text Prompt***
[0049] In the example of the email text prompt shown above, in the picture, "·Title", the press release title is stored, and in "·Article Body", the press release summary is stored. "<Basic Information of the Announcing Company>" is described based on specific basic information. "<Your Company's Basic Information>" is the information of the user (in this case, the news site operator), which has been registered in advance.
[0050] After generating and obtaining the email text prompt, the server information processing unit 10 sends the email text prompt to the large language model server 4 and obtains the business email text information as its response. The business email text information is the information in which the text of the business email is recorded. Here, the email text prompt includes a press release summary that summarizes the gist and key points of the press release, and the press release summary includes proper nouns that are the points of announcement or publicity in the press release. Therefore, the business email text information generated based on the email text prompt is highly suitable as a business email. As described above, the server information processing unit 10 uses the function of the large language model 5 to generate the business email text information. The above is the detailed process of step SA7.
[0051] After step SA7, the server information processing unit 10 executes database registration processing (step SA8). Specifically, the server information processing unit 10 accesses the project management database 37 and registers a new record. Figure 5(B) is a diagram showing the content of one record (hereinafter referred to as "project record") in the project management database 37. As shown in Figure 5(B), the project record has fields such as project ID <f>and the user ID <f>and the original URL <f>and original data <f>and, the title <f>and the main text <f>And, abstract <f>and, basic <f>and the analysis result <f>and the reasons for evaluation <f>and the text <f>and status <f>comprises. The server information processing unit 10 has a case ID <f>and user ID <f>Stores the unique case ID and the user ID of the news site operator. The server information processing unit 10 stores the original URL <f>, Metadata <f>, Title <f>, this article <f>、Abstract <f>, basic <f>Analysis Results <f>, Reasons for evaluation <f>And the text <f>Stores the press release URL, press release data, press release title, press release text, press release summary, specific basic information, analysis result information, evaluation reason information, and sales email text information. The server information processing unit 10 has a status <f>Stores the status management information of the initial value. The status management information will be described later. Hereinafter, a collection of information managed by each record in the case management database 37 is conveniently referred to as a "case". That is, records are registered in the case management database 37 for each case.
[0052] As a result of the above processing, in a specific press release site, each time a new press release is published, the analysis result information is acquired, and a new record including the analysis result information is registered in the case management database 37.
[0053] <Scene S3 where the analysis result information is provided to the news site operator> Next, an operation example of the control system 1 in the scene S3 where the analysis result information is provided to the news site operator will be described. The news site operator starts the browser of the user terminal 3 and performs a predetermined operation to display the main screen on the terminal display unit 15. Next, the news site operator performs a predetermined operation to instruct the display of the case list screen 40. In response to the instruction, the terminal information processing unit 13 displays the case list screen 40 on the terminal display unit 15 under the control of the server information processing unit 10.
[0054] FIG. 8 is a diagram showing an example of the case list screen 40. On the case list screen 40, summary records 41 are displayed in a list for each case (case record) corresponding to a news site operator. The server information processing unit 10 causes each of the summary records 41 to be displayed based on each of the case records corresponding to the news site operator (user ID) in the case management database 37. As shown in FIG. 8, the summary record 41 includes the company name of the source company, the analysis result information, and the current status. The current status means the following. That is, when actually conducting business regarding the advertisement submission to the source company, it will go through various statuses such as the status of not having started business, the status of not having obtained a response after making an inquiry, the status of having obtained a response, the status of being in negotiation, the status of having reached an agreement after negotiation, and the status of not having reached an agreement after negotiation. And the current status means the current status regarding the business for the corresponding source company. The status of the case record <f>The status management information stored therein includes information indicating the current status. The server information processing unit 10 displays a mark indicating that fact (the mark "NEW!" in FIG. 8) on the outline record 41 to be displayed for the first time. The case list screen 40 includes a user interface that can sort the outline records 41 under predetermined conditions or narrow down the outline records 41 to be displayed.
[0055] By referring to the case list screen 40, the news site operator can easily and intuitively grasp the analysis result information for each case related to himself / herself, and can quickly and accurately grasp the fact if there is a company recommended for business. Furthermore, the news site operator can accurately grasp the current status regarding business for each case.
[0056] Each of the outline records 41 is selectable. When an outline record 41 is selected or a detail button displayed beside each record is selected, the server information processing unit 10 displays the case detail screen 42. FIG. 9 is a diagram showing an example of the case detail screen 42. As shown in FIG. 9, the case detail screen 42 includes the basic company information (reference numeral 43) of the corresponding issuing company, the press release title (reference numeral 44), the press release summary (reference numeral 45), and the press release URL (reference numeral 46). By referring to these information, the news site operator can accurately grasp the outline of the issuing company of the corresponding press release and the content of the press release. Also, when the news site operator wants to check the corresponding press release page 30, he / she can check it using the press release URL. Furthermore, the case detail screen 42 includes the analysis result information (reference numeral 47) and the evaluation reason information (reference numeral 48). By referring to these information, the news site operator can grasp not only the degree of recommendation for conducting business, but also why such a degree of recommendation was obtained.
[0057] Furthermore, the project details screen 42 includes a business email text area 49. In the business email text area 49, the business email text information is displayed as text within a frame. The business email text information can be copied and pasted. Also, the business email text area 49 includes a user interface for editing the business email text information and a user interface for downloading the business email text information as data in a predetermined file format. The specific news site operator can copy and paste the business email text information for use in other software, freely edit the business email text information, or download the business email text information. The server information processing unit 10 updates the content of the corresponding business email text information according to the editing.
[0058] Furthermore, the project details screen 42 includes a status details area 50. The status details area 50 includes information indicating the current status and the statuses passed through so far. Also, the status details area 50 includes a user interface for editing the current status and a user interface for editing the statuses passed through so far. The status management information of the project record contains all the information displayed in the status details area 50 or the underlying information thereof. The specific news site operator can freely edit the current status and the statuses passed through so far. The server information processing unit 10 updates the content of the corresponding status management information according to the editing. In this way, the server information processing unit 10 manages the progress of business with the corresponding source company (specific target) by storing and appropriately updating the status management information.
[0059] As described above, the terminal information processing unit 13 has a function of providing the user with the analysis result information, the evaluation reason information, and the business email text information (business article).
[0060] As described above, the information processing system according to the present embodiment acquires specific target person information based on specific public information regarding a specific target person among the public information regarding the disclosed target person, and analyzes the content of the public information from the viewpoint of whether it is recommended to conduct business with the target person regarding a predetermined matter. An information processing unit having a function of acquiring analysis result information indicating the degree to which it is recommended to conduct business with a specific target person is provided based on the analysis information and the specific target person information to be referred to. According to this configuration, based on the specific target person information based on the specific public information disclosed regarding the specific target person, and the "analysis information to be referred to when analyzing the content of the public information from the viewpoint of whether it is recommended to conduct business with the target person regarding a predetermined matter", "analysis result information indicating the degree to which it is recommended to conduct business with the specific target person" is automatically generated. Therefore, the user can recognize to what extent it is recommended to conduct business with the specific target person by using the analysis result information without collecting the information of the specific target person by himself / herself or analyzing the collected information, and can accurately determine whether to conduct business with the specific target person based on this recognition. That is, according to the above configuration, it is possible to enable the user to accurately determine whether to conduct business with a certain target person while reducing the labor of the user.
[0061] As described above, one embodiment of the present invention has been described. However, the above embodiment is merely an example of a specific implementation when implementing the present invention, and the technical scope of the present invention should not be construed in a limited manner by this. That is, the present invention can be implemented in various forms without departing from the gist or its main features. Hereinafter, modified examples of the above embodiment will be shown. The following modified examples may be applied in combination if they can be combined.
[0062] · In the above embodiment, the server information processing unit 10 acquires the evaluation reason information together with the analysis result information in the analysis process, but the configuration may be such that the evaluation reason information is not acquired. · In the above embodiment, the sales activity was related to "publishing advertisements on news sites". That is, the "predetermined matter" was "publishing advertisements on services where advertisement provision is carried out". However, what the sales activity is about is not limited. · In the above embodiment, the analysis result information was information indicating the degree of recommendation for conducting sales activities on a five - level evaluation. However, the content of the analysis result information is not limited to the exemplified content. That is, the analysis result information may be information indicating the degree of recommendation for conducting sales activities. For example, regarding the five - level evaluation, it may be a three - level evaluation or a ten - level evaluation. Also, the analysis result information may be information indicating whether to proceed with sales activities, such as "should conduct sales / should not conduct sales". Further, the analysis result information may be a score / points that takes values within a range from the minimum value to the maximum value. Also, the analysis result information may be information expressing the degree of recommendation for conducting sales activities as a text.
[0063] · In the above embodiment, the analysis information was positive element information and negative element information. However, the content of the analysis information is not limited to the exemplified content. For example, the analysis information may include conditions that are essential for determining the counterpart for conducting sales activities, conditions that definitely determine the counterpart for conducting sales activities, or conditions that definitely do not determine the counterpart for conducting sales activities. Among these conditions, the "conditions that definitely determine the counterpart for conducting sales activities" is, for example, the condition that "the content of the press release is an introduction of a service using ○○ of the conversational AI". · In the above embodiment, the analysis information was set by the user. Regarding this point, a configuration where a person other than the user sets the analysis information may also be possible. Also, a configuration where the analysis information is set by automatic means may be possible. In this configuration, for example, the server information processing unit 10 uses information related to the user and utilizes the large - scale language model 5 to generate and acquire the analysis information. · In the above embodiment, the analysis information was input and set as text. Regarding this point, a configuration where it is selected and set from a plurality of pre - prepared information may also be possible.
[0064] · In the above embodiment, the server information processing unit 10 detected that a new press release was published on a specific press release site and acquired the new press release. In this regard, the server information processing unit 10 may be configured to periodically access a specific press release to acquire the press release. Also, the server information processing unit 10 may be configured to passively acquire a press release by using a service that distributes press releases. · In the above embodiment, the server information processing unit 10 acquired publicly available information (press release) published on a specific site / service. In this regard, the server information processing unit 10 may be configured to widely target media published on the Internet to acquire publicly available information. In the case of this configuration, for example, the server information processing unit 10 uses existing technologies related to crawling and scraping to acquire press releases that include or are likely to include information about persons who may be business counterparts. · In the above embodiment, the server information processing unit 10 targeted all acquired press releases for processing. In this regard, when the server information processing unit 10 acquires a press release, it may be configured to determine whether to target it for processing, and discard those determined not to be targeted for processing. · In the above embodiment, the server information processing unit 10 directly accessed the press release page 30 via the Internet to acquire press release data. In this regard, the press release data may be registered in a predetermined database, and the server information processing unit 10 may be configured to access the predetermined database and acquire press release data from the predetermined database. In this configuration, the registration of the press release data in the predetermined database may be performed including artificial means. Also, a configuration may be adopted in which a user provides a press release (or information that may substitute for this) to the server information processing unit 10.
[0065] · In the above-described embodiment, the server information processing unit 10 acquired the press release as public information. Regarding this point, the public information is not limited to press releases. For example, the public information may be information published / advertised by the target person on their own website or other media. Also, for example, the public information may be an advertisement regarding the target person. Also, for example, the public information may be news or an introduction regarding the target person provided by news agencies, mass media, or other parties. · In the above-described embodiment, the server information processing unit 10 acquired public information (press release) published on the Internet. Regarding this point, the public information does not have to be information published on the Internet. For example, the public information may be information published in newspapers, magazines, on television, on the radio, or by other media.
[0066] · In the above-described embodiment, the press release title, press release summary, and specific basic information corresponded to the specific target person information. However, the content of the specific target person information is not limited to the exemplified content. For example, the configuration may be such that the press release title or specific basic information is not included in the specific target person information. Also, for example, the configuration may be such that the press release text is included in the specific target person information instead of or together with the press release summary. Also, for example, if the large language model 5 is supported, the configuration may be such that still images, videos, or audio included in the press release are included in the specific target person information. · In the above-described embodiment, the press release title and press release text corresponded to the "related article", and based on this, the press release summary was generated. However, the related article is not limited to the exemplified information. For example, only the press release text may be used. · In the above-described embodiment, the server information processing unit 10 generated and acquired the specific target person information based on the press release. Regarding this point, a configuration may be such that the specific target person information is generated in advance and registered in a predetermined database, and the server information processing unit 10 acquires the specific target person information from the predetermined database. In this configuration, the generation and registration of the specific target person information may be performed including artificial means.
[0067] · In the above-described embodiment, the server information processing unit 10 generated the business email text information for all of the press releases that were the basis of the analysis result information, regardless of the value of the analysis result information. In this regard, the server information processing unit 10 may be configured to obtain the business email text information only when the value of the analysis result information is equal to or greater than the threshold value. In this configuration, the server information processing unit 10 may be configured to register a record in the project management database 37 only when the value of the analysis result information is equal to or greater than the threshold value. Also, in this regard, the server information processing unit 10 may be configured to obtain the business email text generation information only when there is an instruction from the user.
[0068] · In the above-described embodiment, the large language model 5 was installed on the large language model server 4 connected to the network N. In this regard, the large language model 5 may be installed in the server storage unit 12 of the information processing server 2. · In the above-described embodiment, the server information processing unit 10 executed the analysis process using the large language model 5. In this regard, the server information processing unit 10 may execute the analysis process without using the large language model 5. Hereinafter, an example of the analysis process without using the large language model 5 is shown. FIG. 10 is a diagram used for explaining the example. First, learning data including at least a combination of analysis information and specific target person information subjected to morphological analysis, vectorization, and other preprocessing, and analysis result information as correct answer data is sufficiently prepared, and a machine learning model is learned using the learning data. The machine learning model is a model that takes the preprocessed analysis information and specific target person information as input and outputs the analysis result information. Hereinafter, the learned machine learning model is referred to as the "learned model". Then, as shown in FIG. 10, in the analysis process, the server information processing unit 10 inputs the preprocessed analysis information and specific target person information into the learned model and obtains the output analysis result information. The above is the example. In this example, the machine learning model may be configured to output evaluation reason information in addition to the analysis result information. In the case of this configuration, for example, as information indicating the reason for evaluation, a plurality of individual reason information is prepared in consideration of the type / typicality of the reason. And the machine learning model is a model that outputs, as the evaluation reason information, the information with the highest possibility of corresponding among the plurality of prepared individual reason information (it may be the possibility of correspondence for each of the individual reason information).
[0069] · In the above-described embodiment, the server information processing unit 10 obtained information indicating the text to be described in the email during business as business text. However, the content of the business text is not limited to the exemplified content. For example, the business text may be configured to generate "information indicating the text of the information to be input into the inquiry form" or "information indicating the text to be conveyed at the telephone call when conducting business by phone". · Regarding the above-described embodiment, a configuration may be adopted in which part or all of the data stored in the server storage unit 12 is stored in the storage unit of an external device accessible by the server information processing unit 10. · The functional blocks shown in the above embodiments can be realized by any hardware or the cooperation of any hardware and any software. That is, these functional blocks are not limited to specific hardware. · Further, the provision of a program executed by a computer of the information processing server 2 or the user terminal 3 can be included in the embodiment. Further, the provision of a recording medium in which the program is recorded in a computer-readable manner can be included in the embodiment. As the above recording medium, a magnetic, optical recording medium or a semiconductor memory device can be used. Specifically, portable or fixed recording media such as flexible disks, HDDs (Hard Disk Drives), CD-ROMs (Compact Disk Read Only Memories), DVDs (Digital Versatile Disks), Blu-ray (registered trademark) Discs, magneto-optical disks, flash memories, and card-type recording media can be mentioned.
[0070] · Also, in the above embodiment, the device having one function is not limited to the device exemplified as the device having the one function. For example, focus on the function of acquiring analysis result information (referred to as the "first function") and the function of providing the analysis result information to the user (referred to as the "second function"). In the above embodiment, as shown in FIG. 11(A), the server information processing unit 10 of the information processing server 2 has the first function, and the terminal information processing unit 13 of the user terminal 3 has the second function. In this configuration, the combination of the information processing server 2 and the user terminal 3 functions as an "information processing system". Also, the server information processing unit 10 and the terminal information processing unit 13 function as "information processing units", respectively. On the other hand, as shown in FIG. 11(B), the terminal information processing unit 13 of the user terminal 3 may have the first function and the second function. In this configuration, the user terminal 3 functions as an "information processing system", and the terminal information processing unit 13 functions as an "information processing unit". Also, as shown in FIG. 11(C), the server information processing unit 10 of the server information processing unit 10 may have the first function and the second function. Note that when the server information processing unit 10 and the terminal information processing unit 13 each function as an "information processing unit", the functions provided in each information processing unit are not limited. Also, a configuration in which one function is realized by the cooperation of the server information processing unit 10 and the terminal information processing unit 13 may be used. Further, the device that realizes a certain function may be a device other than the information processing server 2 and the user terminal 3.
Explanation of Signs
[0071] 1 Control System (Information Processing System) 2 Information Processing Server (Information Processing System) 3 User Terminal (Information Processing System) 4 Large Language Model Server 5 Large Language Model 10 Server Information Processing Unit (Information Processing Unit) 13 Terminal Information Processing Unit (Information Processing Unit)< / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f> < / f>
Claims
1. A function for acquiring specific target information based on specific public information on specific targets from among public information on targets that has been made public, and acquiring analytical information to be referenced when analyzing the content of the public information from the perspective of whether or not it is recommended to do business with the target in relation to a specified matter, and acquiring analysis result information indicating the degree to which it is recommended to do business with the specific target based on the specific target information; an information processing unit having a function of acquiring a first prompt including the analysis information and the specific target information, the first prompt requesting that the specific target information be analyzed by reflecting the analysis information to generate the analysis result information, providing the first prompt to a language model, and acquiring a response sentence from the language model as the analysis result information; The specific public information includes announcements or advertisements by the specific target person, or related articles including news or introductions about the specific target person, The information processing unit includes: obtaining a second prompt including the related article and requesting generation of a summary of the related article, providing the second prompt to the language model, and obtaining a response sentence from the language model as the summary; and acquiring the first prompt including the summary as the specific target information. An information processing system comprising:
2. The information processing unit includes: and obtaining the second prompt, the second prompt requesting that the summary be generated such that proper nouns present in the related article are included in the summary.
2. The information processing system according to claim 1 .
3. A function for acquiring specific target information based on specific public information on specific targets from among public information on targets that has been made public, and acquiring analytical information to be referenced when analyzing the content of the public information from the perspective of whether or not it is recommended to conduct business with the target in relation to a specified matter, and acquiring analysis result information indicating the degree to which it is recommended to conduct business with the specific target based on the specific target information; a function of acquiring a first prompt, the first prompt including the analysis information and the specific target information, requesting that the specific target information be analyzed by reflecting the analysis information to generate the analysis result information, providing the first prompt to a language model, and acquiring a response sentence from the language model as the analysis result information; a function of acquiring the first prompt for requesting generation of evaluation reason information indicating the reason why the analysis result information has become the value thereof in addition to the analysis result information; an information processing unit having a function of providing the analysis result information and the evaluation reason information to a user; The information processing system according to the present invention is characterized in that
4. An information processing unit having a function of acquiring specific target information based on specific public information on specific targets from among public information on targets that has been made public, and acquiring analytical information to be referred to when analyzing the content of the public information from the perspective of whether or not it is recommended to do business with the target in relation to a specified matter, and acquiring analysis result information indicating the degree to which it is recommended to do business with the specific target based on the specific target information, The analysis information includes information indicating factors that are factors that will lead to a positive judgment regarding conducting business with the target person, and information indicating factors that will lead to a negative judgment. An information processing system comprising:
5. An information processing unit having a function of acquiring specific target information based on specific public information on specific targets from among public information on targets that has been made public, and acquiring analytical information to be referred to when analyzing the content of the public information from the perspective of whether or not it is recommended to do business with the target in relation to a specified matter, and acquiring analysis result information indicating the degree to which it is recommended to do business with the specific target based on the specific target information, The predetermined matters include placing advertisements on the service for which advertisements are provided. An information processing system comprising:
6. The information processing unit includes: a function of acquiring a third prompt including the specific target information and requesting the target to generate sales text that can be used when the user makes sales by reflecting the specific target information, providing the third prompt to the language model, and acquiring a response sentence from the language model as the sales text; The function of providing the sales document to a user.
4. The information processing system according to claim 1, wherein the information processing system comprises: a first input section;
7. The information processing unit has a function of providing the analysis result information to a user.
6. The information processing system according to claim 1, wherein the information processing system comprises:
8. The information processing unit includes: a function of providing a user interface into which the analysis information can be input; and acquiring the analysis information based on information input to the user interface.
6. The information processing system according to claim 1, wherein the information processing system comprises:
9. The information processing unit has a function of managing the progress of sales to the specific target person.
6. The information processing system according to claim 1, wherein the information processing system comprises:
10. The information processing unit has a function of acquiring the public information published on the Internet.
6. The information processing system according to claim 1, wherein the information processing system comprises:
11. The public information includes press releases about the subject published on the Internet.
11. The information processing system according to claim 10.
12. An information processing unit of the information processing system acquires specific target information based on specific public information on a specific target from public information on the target that has been made public; The information processing unit of the information processing system acquires analysis result information indicating the degree to which it is recommended to conduct business with the specific target based on analysis information to be referred to when analyzing the content of public information from the viewpoint of whether or not it is recommended to conduct business with the target in a predetermined matter and the specific target information, The information processing unit includes: a function of acquiring a first prompt, the first prompt including the analysis information and the specific target information, requesting that the specific target information be analyzed by reflecting the analysis information to generate the analysis result information, providing the first prompt to a language model, and acquiring a response sentence from the language model as the analysis result information; The specific public information includes announcements or advertisements by the specific target person, or related articles including news or introductions about the specific target person, The information processing unit includes: obtaining a second prompt including the related article and requesting generation of a summary of the related article, providing the second prompt to the language model, and obtaining a response sentence from the language model as the summary; and acquiring the first prompt including the summary as the specific target information.
23. An information processing method comprising:
13. A step in which an information processing unit of an information processing system acquires specific target information based on specific public information regarding a specific target from public information regarding the target that has been made public; The information processing unit of the information processing system acquires analysis result information indicating the degree to which it is recommended to conduct business with the specific target based on analysis information to be referred to when analyzing the content of public information from the viewpoint of whether or not it is recommended to conduct business with the target in a predetermined matter and the specific target information, The information processing unit includes: a function of acquiring a first prompt, the first prompt including the analysis information and the specific target information, requesting that the specific target information be analyzed by reflecting the analysis information to generate the analysis result information, providing the first prompt to a language model, and acquiring a response sentence from the language model as the analysis result information; a function of acquiring the first prompt for requesting generation of evaluation reason information indicating the reason why the analysis result information has become the value thereof in addition to the analysis result information; and providing the analysis result information and the evaluation reason information to a user.
23. An information processing method comprising:
14. A step in which an information processing unit of an information processing system acquires specific target information based on specific public information regarding a specific target from public information regarding the target that has been made public; The information processing unit of the information processing system acquires analysis result information indicating the degree to which it is recommended to conduct business with the specific target based on analysis information to be referred to when analyzing the content of public information from the viewpoint of whether or not it is recommended to conduct business with the target in a predetermined matter and the specific target information, The analysis information includes information indicating factors that are factors that will lead to a positive judgment regarding conducting business with the target person, and information indicating factors that will lead to a negative judgment.
23. An information processing method comprising:
15. A step in which an information processing unit of an information processing system acquires specific target information based on specific public information regarding a specific target from public information regarding the target that has been made public; The information processing unit of the information processing system acquires analysis result information indicating the degree to which it is recommended to conduct business with the specific target based on analysis information to be referred to when analyzing the content of public information from the viewpoint of whether or not it is recommended to conduct business with the target in a predetermined matter and the specific target information, The predetermined matters include placing advertisements on the service for which advertisements are provided.
23. An information processing method comprising:
Citation Information
Patent Citations
Information providing system, information providing method, and program
JP2021131843A
Information processing system
JP2021149966A
Marketing auto-assisting device, marketing auto-assisting system, and marketing auto-assisting method
JP2022084059A
Sales support device and sales support method
JP2024016411A
Method and appratus for analyzing investment target company
KR1020220042864A