Sentence creating program and information processing device

Abstract

To provide a sentence creating program and an information processing device that create a sentence in accordance with a user at a transmission destination and with a transmission purpose.SOLUTION: An information processing device 1 includes: user selecting means 100 for selecting a user who is the transmission destination of a scouting text; recruit selecting means 101 for selecting recruit information 112 that is prepared in advance; attraction extraction requesting means 102 for requesting a generative AI2 to extract a plurality of attractions from the selected user information 111 on the user; and sentence creation requesting means 103 for understanding the recruit information 112 prepared in advance, and for requesting the generative AI2 to create a sentence for the user based on the plurality of extracted attractions.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a text generation program and an information processing apparatus.

Background Art

[0002] As a conventional technique, an information processing apparatus that reuses a previously used document when creating a stereotyped document has been proposed (see, for example, Patent Document 1).

[0003] The information processing apparatus disclosed in Patent Document 1 includes a message storage unit for storing transmitted and received messages, a selection condition input unit for inputting selection conditions including the presence or absence of a reuse relationship to acquire a message from this message storage unit, a message selection unit for selecting a desired message from the message storage unit using the selection conditions input from this selection condition input unit, a message creation unit for displaying the message selected by this message selection unit as a reuse source message and creating a new message, and a message reuse relationship storage unit for storing the correspondence between the new message created by this message creation unit and the reuse source message. The information processing apparatus of Patent Document 1, with this configuration, when creating a stereotyped document, searches for a template document from the stored documents and creates a desired document by editing this template document.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] According to the information processing apparatus described above, when creating a typical document, it searches for a template document from the accumulated documents and creates a desired document by editing this template document. However, for example, when creating a message for recruitment after finding the necessary human resources, there is a problem that a message corresponding to the characteristics of the recipient user and the purpose of transmission cannot be created.

[0006] An object of the present invention is to provide a sentence generation program and an information processing apparatus that create a sentence corresponding to the recipient user and the purpose of transmission.

Means for Solving the Problem

[0007] One aspect of the present invention provides the following sentence generation program and information processing apparatus in order to achieve the above object.

[0008] [1] A computer is a selection means for receiving a selection of a user who is the recipient of the scout message and prepared recruitment information, an extraction request means for requesting a generation AI to extract a plurality of attractions from the user information of the user, A sentence generation program for functioning as a generation request means for requesting the generation AI to generate a sentence for the user based on the plurality of attractions extracted by understanding the prepared recruitment information. [2] A computer is The sentence generation program according to [1] above, which further functions as a generation means for generating a scout message using at least one of the sentences for the user generated by the generation AI and a prepared standard sentence. [3] The generation request means requests the generation AI to generate the sentence for types other than the types of the prepared standard sentences. The sentence generation program according to [2] above. [4] The generation request means learns the past sentences of the user of the transmission source and requests the generation AI to make the sentence in the style of the user of the transmission source. The sentence generation program according to any one of [1] to [3] above. [5] When the generation request means requests the generation of the sentence, it requests the generation AI as described in any one of [1] to [3] above to touch on the common carrier of the users who are common friends of the user of the source and the user of the destination so as to generate the sentence. [6] When the generation request means requests the generation AI to learn sentence generation, it targets scout sentences with a high reply rate and requests the generation AI as described in any one of [1] to [3] above to generate sentences. [7] A user who is the destination of a scout sentence, a selection means for receiving a selection of recruitment information prepared in advance, An extraction request means for requesting the generation AI to extract a plurality of attractions from the user information of the user, An information processing apparatus having a generation request means for understanding the recruitment information prepared in advance and requesting the generation AI to generate a sentence for the user based on the plurality of attractions extracted. [8] A step of receiving a selection of recruitment information prepared in advance from a user who is the destination of a scout sentence, A step of requesting the generation AI to extract a plurality of attractions from the user information of the user, A sentence generation method having a step of requesting the generation AI to generate a sentence for the user based on the plurality of attractions extracted by understanding the recruitment information prepared in advance. [9] A generation AI that generates at least sentences in response to requests, A user who is the destination of a scout sentence, a selection means for receiving a selection of recruitment information prepared in advance, an extraction request means for requesting the generation AI to extract a plurality of attractions from the user information of the user, and a generation request means for understanding the recruitment information prepared in advance and requesting the generation AI to generate a sentence for the user based on the plurality of attractions extracted. An information processing system comprising an information processing apparatus having the same.

Effect of the Invention

[0009] According to the inventions according to claims 1, 7, 8, and 9, sentences corresponding to the destination user and the transmission purpose can be created. According to the invention according to claim 2, a scout sentence can be generated using at least one of the sentences for the user generated by the generative AI and the prepared fixed-form sentences. According to the invention according to claim 3, it is possible to request the generative AI to generate sentences for types other than the types of the prepared fixed-form sentences. According to the invention according to claim 4, it is possible to learn the past sentences of the user of the sender and request the generative AI to make the sentences in the style of the user of the sender. According to the invention according to claim 5, when requesting the generation of a sentence, it is possible to request the generative AI to touch on the common carrier of the user who is a common friend of the user of the sender and the user of the recipient. According to the invention according to claim 6, it is possible to request the pre-AI to learn sentence generation for scout sentences with a high reply rate.

Brief Description of the Drawings

[0010]

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

[0011] [Embodiment] (Configuration of Information Processing System) FIG. 1 is a schematic diagram showing an example of the configuration of an information processing system according to an embodiment.

[0012] This information processing system is configured by connecting an information processing apparatus 1, a generation AI 2, and terminals 3a, 3b,... to be communicable with each other via a network 4. The terminals 3a, 3b,... are each operated by users 5a, 5b,....

[0013] The information processing apparatus 1 is a server-type information processing apparatus that operates in response to requests from the terminals 3a, 3b,... and includes electronic components such as a CPU (Central Processing Unit) and a flash memory having functions for processing information within the main body.

[0014] The generation AI 2 is a server-type information processing apparatus that mainly functions as a generation AI that learns a neural network and generates a sentence or the like for a certain input, operates in response to requests from the information processing apparatus 1, and includes electronic components such as a CPU, a GPU (Graphics Processing Unit), and a flash memory having functions for processing information within the main body. The generation AI 2 can use generally provided services, and for example, it may use Azure OpenAI Service, OpenAI API, etc.

[0015] The terminals 3a, 3b,... are information processing apparatuses such as PCs (Personal Computers) and include electronic components such as a CPU and a flash memory having functions for processing information within the main body.

[0016] The network 4 is a communication network capable of high-speed communication, for example, a wired or wireless communication network such as an intranet or a LAN (Local Area Network).

[0017] The information processing apparatus 1, as an example, generates a text (hereinafter sometimes referred to as a "scouting text") for scouting the user 5a in response to a request from the terminal 3b. Specifically, when the user 5a who is the scouting target and a certain recruitment information are selected in the terminal 3b in cooperation with the generation AI 2, the information processing apparatus 1 generates a text for scouting the user 5a according to the content of the recruitment information.

[0018] (Configuration of the information processing apparatus) FIG. 2 is a block diagram showing a configuration example of the information processing apparatus 1 according to the embodiment.

[0019] The information processing apparatus 1 includes a control unit 10 composed of a CPU or the like, which controls each part and executes various programs, a storage unit 11 composed of a storage medium such as a flash memory for storing information, and a communication unit 12 for communicating with the outside via the network 4.

[0020] By executing the text generation program 110 described later, the control unit 10 functions as a user selection means 100, a recruitment selection means 101, an attraction extraction request means 102, a text generation request means 103, a text generation means 104, and the like.

[0021] The user selection means 100 selects a user who is the destination of the scouting text as the scouting target from the users in the user information 111.

[0022] The recruitment selection means 101 selects a job recruitment for scouting from the recruitment information 112.

[0023] The attraction extraction request means 102 requests the generation AI 2 to extract, for example, three attractions of the user based on the information of the user selected by the user selection means 100 from the user information 111.

[0024] The sentence generation requirement means 103 requests the generation AI 2 to generate a sentence for the user based on understanding the content of the recruitment information 112 and the three attractions extracted by the attraction extraction requirement means 102.

[0025] The sentence generation means 104 performs display processing on the generation result generated by the generation AI 2 on the terminals 3a, 3b..., and generates scout sentence information 114 based on the operation content received for the display result and stores it in the storage unit 11.

[0026] The storage unit 11 stores a sentence generation program 110 that operates the control unit 10 as the above-described respective means 100 - 104, user information 111, recruitment information 112, fixed-form sentence information 113, scout sentence information 114, and the like.

[0027] Figure 3 is a schematic diagram showing a configuration example of the user information 111.

[0028] The user information 111 includes a user ID for identifying the user, a user name, an affiliation indicating the company to which the user belongs, an occupation, a self-introduction sentence, what can be done, a career which is the work history, a participating community indicating the community in which the user participates, and the like.

[0029] Figure 4 is a schematic diagram showing a configuration example of the recruitment information 112.

[0030] The recruitment information 112 has a recruitment ID for identifying the recruitment, a title, a category, a target, and content. The category is a category regarding the purpose of the recruitment and has items such as a formal "member recruitment" and a "casual interview" where participation is possible even without the intention of changing jobs.

[0031] Figure 5 is a schematic diagram showing a configuration example of the fixed-form sentence information 113.

[0032] The fixed-form sentence information 113 has a fixed-form sentence ID for identifying the fixed-form sentence, a title, and content.

[0033] (Operation of the information processing apparatus) Next, the operations of this embodiment will be described by dividing them into (1) preparation operations, (2) scout message creation operations, and (3) fixed-form message registration and editing operations.

[0034] (1) Preparation operations First, the user 5a, who is a job seeker, operates the terminal 3a to input information such as the user name, affiliation, occupation, self-introduction, skills, career history, and participating communities for user registration in the service provided by the information processing apparatus 1. The input information is transmitted from the terminal 3a to the information processing apparatus 1 and recorded as user information 111 in the storage unit 11.

[0035] Also, the user 5b, who wants to scout for human resources, operates the terminal 3b to similarly register as a user in the service provided by the information processing apparatus 1, and then operates the terminal 3b to register information such as the title, category, target, and content as the recruitment details. The input information is transmitted from the terminal 3b to the information processing apparatus 1 and recorded as user information 111 and recruitment information 112 in the storage unit 11, respectively.

[0036] Also, when scouting for human resources by operating the terminal 3b, the user 5b registers fixed-form messages to be used in the scout messages to be transmitted. The input information is transmitted from the terminal 3b to the information processing apparatus 1 and recorded as fixed-form message information 113 in the storage unit 11. Note that the fixed-form message information 113 registered here is a message other than the message generated by the generation AI 2, such as an introductory greeting, company introduction, interview schedule adjustment, and closing greeting. The details of the fixed-form message registration operation will be described in "(3) Fixed-form message registration and editing operations".

[0037] (2) Scout message creation operations Next, the user 5b operates the terminal 3b to view the information of registered users via the service provided by the information processing apparatus 1. Hereinafter, the situation where the user 5b scouts the user 5a for the occupation the user 5b is recruiting for will be described.

[0038] User 5b first operates terminal 3b to request the display of a screen for creating a scout message. When a request is sent from terminal 3b to information processing device 1, the screen for creating a scout message is display-controlled on terminal 3b. While checking the display screen of terminal 3b, user 5b operates terminal 3b to select a user who will be the recipient of the scout message.

[0039] Figure 11 is a flowchart for explaining the scout message creation operation.

[0040] The user selection means 100 of information processing device 1 selects a user who will be the recipient of the scout message as the scouting target from the users in user information 111 according to the operation content of terminal 3b (S1). For example, the case of selecting "Masahiro Yamada" with user ID "V003" in user information 111 in FIG. 3 will be described.

[0041] Next, the charm extraction request means 102 requests the generation AI2 to extract, for example, three charms of the user based on the information (profile) of the user selected by the user selection means 100 from user information 111 (S2). Here, the profile is assumed to include the affiliation, occupation, self-introduction text, skills, career, and participating communities of user information 111.

[0042] The charm extraction request means 102 requests the generation AI2 as follows, for example. "First, please read the following profile text. ==========Profile Information of Mr. Yamada========== ■ Company of the other party METRUST Co., Ltd. ■ Occupation of the other party Engineering / Web Engineer ■ Self-introduction text of the other party I am Yamada from METRUST. After working as a server-side engineer at DNE Co., Ltd. for about 6 years, I was invited by the representative Mr. Iwasaki and participated in the establishment of YOUTRUST. I'm working hard on the development every day! If you have any requests, bugs, or need help, please feel free to send a message! P.S. I like poker, so I'm also looking for people who can play poker together! Laugh ■ What the other party can do AWS React Ruby on Rails ■ The other party's career · Career 1 Company: METRUST Co., Ltd. Start date: April 1, 2018 00:00:00 +0900 End date: Project / Department name: Web application development Role: Engineer Job content: I'm working really hard! https: / / metrust.jp / · Career 2 Company: DNE Co., Ltd. Start date: July 1, 2012 00:00:00 +0900 End date: September 1, 2013 00:00:00 +0900 Project / Department name: Sengoku Royale Role: Engineer Job content: It was the first team I was assigned to after the new graduate training. I used the in-house FW called MobaSiF and developed and operated it vigorously in Perl. · Career 3 Company: DNE Co., Ltd. Start date: October 1, 2013 00:00:00 +0900 End date: June 1, 2014 00:00:00 +0900 Project / Department Name: Gundam Royal Role: Engineer Job Content: Lead the development of various measures and new events. Received the in - department MVP individually and the President's Award as a team. ·Career 4 Company: D.N.A., Inc. Start Date: 2014 - 07 - 01 00:00:00 +0900 End Date: 2018 - 03 - 01 00:00:00 +0900 Project / Department Name: Sakasho Role: Engineer Job Content: Engage in the development of Mobile BaaS. Reference: https: / / www.slideshare.net / blueskyblue / dena - sakasho ■Community where the other party participates Engineers' Association Web Startup Engineers Re - learning for working adults Board game lovers, gather! ========== Yamada Masahiro | Please summarize the three attractions of METRUST. Please follow the output format below. {number} 847f9dda - 8afa - 4ffa - b646 - 88f8a8df5956 {attractive point title} 847f9dda - 8afa - 4ffa - b646 - 88f8a8df5956 {attractive point content} {number} 847f9dda-8afa-4ffa-b646-88f8a8df5956 {Attractive Point Title} 847f9dda-8afa-4ffa-b646-88f8a8df5956 {Attractive Point Content} {number} 847f9dda-8afa-4ffa-b646-88f8a8df5956 {Attractive Point Title} 847f9dda-8afa-4ffa-b646-88f8a8df5956 {Attractive Point Content} ... ■Points to Note ·{Attractive Point Title} should be within 30 characters ·{Attractive Point Content} should be within 200 characters ·Please add "san" to the company name. (Do not add it to YOUTRUST Co., Ltd.) Then, please

[0043] Based on the above requirements, Generation AI2 analyzes the affiliation, job type, self-introduction text, capabilities, career, participating communities, etc. of User Information 111 and generates three attractions. For example, it generates "broad career experience", "communication skills", and "technical skills" as attractions.

[0044] After the above operation or in parallel with the above operation, the text generation means 104 performs a display process on the following screen on the display unit of the terminal 3b. The terminal 3b displays the following screen 30a on the display unit.

[0045] Figure 6 is a schematic diagram showing a configuration example of the screen displayed on the display units of terminals 3a and 3b.

[0046] Screen 30a is a screen for determining how to address the user who is the recipient of the scout message, and has the recipient's user name 300a, a call method input field 301a, and a Next button 302a for determining the call method and transitioning to the next screen.

[0047] The user 5b checks the content of the screen 30a, inputs a desired calling method into the calling method input field 301a via the terminal 3b, and presses the Next button 302a.

[0048] The text generation means 104 receives the calling method from the terminal 3b (S3).

[0049] Next, in order to allow the recruitment to be selected, the text generation means 104 performs a display process on the display unit of the terminal 3b for the following screen. The terminal 3b displays the following screen 30b on the display unit.

[0050] FIG. 7 is a schematic diagram showing a configuration example of the screens displayed on the display units of the terminals 3a and 3b.

[0051] The screen 30b is a screen for selecting a recruitment for scouting, and has a display column 300b in which a list of recruitment information created by the user 5b is displayed, a selection button 301b for selecting the recruitment, and a skip button 302b for transitioning to the next screen without selecting the recruitment.

[0052] The user 5b checks the content of the screen 30b, selects a desired recruitment via the terminal 3b, and presses the selection button 301b.

[0053] The recruitment selection means 101 selects a job recruitment for scouting from the recruitment information 112 based on the operation content on the terminal 3b (S4). For example, the case where "Great recruitment for YOUTRUST engineers!" with the recruitment ID "R003" in the recruitment information 112 in FIG. 4 is selected will be described.

[0054] Next, the text generation request means 103 understands the content of the selected recruitment information 112 and requests the generation AI2 to generate a sentence for the user based on the three attractions extracted by the attraction extraction means 102 (S5).

[0055] The text generation request means 103 requests the generation AI2 as follows, for example. Please read and understand the recruitment information first. ==========Recruitment Information========== Title: YOUTRUST is urgently recruiting engineers! Text: YOUTRUST is currently recruiting web engineers to develop two products (Career SNS / HR Tech Saas)! Those who are interested in the development at YOUTRUST, please feel free to have a casual conversation starting from an informal interview! If you seem like a good fit, we will contact you from our company. [Required Skills] - More than 3 years of development practical experience using an MVC framework with LL languages such as Ruby, PHP, Python - DB design and operation experience considering normalization and indexing in MySQL - Operation experience of public cloud services such as AWS or GCP - Team development experience with multiple people using GitHub [Welcome Skills] - Front-end development experience - Development and operation experience of web APIs (REST, GraphQL, gRPC) for native apps or front-ends - Development experience using Ruby on Rails - Experience in building development processes such as CI and automated testing - Experience in product management - Development experience in large-scale services - Experience in large-scale refactoring - In-depth knowledge of QA - In-depth knowledge of security You are Mr. Sato Haruto, a recruiter from the company "YOUTRUST Co., Ltd.". You are about to send a scouting message to Mr. Masahiro Yamada. ■ Regarding the recruitment message One of the attractive points of Mr. Masahiro Yamada is his "communication skills". Mr. Yamada, who is also a member of the "Engineers' Club", values communication among engineers. He learns new knowledge and applies it to his work. Please create a message that touches on this point. ■ Precautions When creating the recruitment message, please pay attention to the following points. · Keep the amount of text within about `100 - 150 characters`. · Avoid making the writing style too rigid. · Do not make the content too pushy. · Try to refrain from using an assertive tone as much as possible. Use expressions like "I thought that ~~". · Use honorific language. (such as "I thought", "I felt", "was", "did", etc.) · Add "san" to the company name. (Do not add it to YOUTRUST Co., Ltd.) ■ NG words that should not be used · You ■ Regarding the output format of the recruitment message The general composition of a common recruitment is as follows. 1. {{Introduction greeting}} 2. {{Company description}} 3. {{Personalized message}} 4. {{Interview appointment arrangement}} 5. {{Closing greeting}} Please make the output content only for the third part, {{Personalized message}}, among the above. ■ Regarding the composition of the {{Personalized message}} in the recruitment message Also, please touch on the image of the talent sought by your company from the recruitment information. As for the structure, it is good to first mention the problems our company is facing and the type of talent we are looking for, Then, in response to that, say that you are looking for someone with a career like Yamada Masahiro. Then, based on the other person's profile, please output only the {{personalized text}} part in about 100 to 150 characters. "

[0056] The text generation request means 103 requests that a text be generated for each attraction, but it may also request that text be generated for multiple attractions.

[0057] The generation AI2 generates text for users to be scouted in accordance with the contents of the recruitment in response to the request. Specifically, three sentences are generated corresponding to the three attractions.

[0058] The sentence generation means 104 receives the generation result generated by the generation AI2 (S6), and processes the screen shown below to display on the terminals 3a, 3b, etc.

[0059] Figure 8 is a schematic diagram showing an example of the configuration of the screen displayed on the display unit of the terminals 3a and 3b.

[0060] Screen 30c is a screen for creating a scouting text, and has an area 300c for copying and inputting text, an area 301c for displaying template text, an area 302c for displaying the contents of the selected recruitment, an area 303c for displaying the text generated by the generation AI2, a close button 304c for canceling the creation of the scouting text, and a content confirmation button 305c for confirming the contents copied and input in area 300c. It also has a management button 301c' for managing template text, a recruitment confirmation button 302c' for confirming the contents of the selected recruitment in a separate window, and a profile confirmation button 303c' for confirming the profile of the user to whom the text is to be sent in a separate window.

[0061] The user 5b checks the content on the screen 30c, selects a desired snippet from the sentence candidates (hereinafter sometimes referred to as "snippets") displayed in the areas 301c, 302c, and 303c via the terminal 3b, and copies and inputs it into the area 300c.

[0062] In addition, for example, when the scout sentence to be created selects the first greeting, company introduction, a snippet that focuses on "extensive career experience" among the snippets for Mr. Imai, and a closing greeting, it will be as follows. "Mr. Yamada" Nice to meet you. I'm Sato from YOUTRUST Co., Ltd. We are enthusiastically engaged in the development of sustainable energy solutions. We welcome those who can drive innovative projects by making the most of their knowledge and skills in thermodynamics for the development of next-generation energy technologies. Our company is looking for engineers with different perspectives. Mr. Yamada's experience at DNE and his activities as a web engineer at METRUST Co., Ltd. exactly match the type of talent we are looking for. If possible, could we have some time for a casual interview? That's all. Sorry to trouble you while you're busy, but we look forward to your reply.

[0063] The user 5b checks the content input in the area 300c, and if the content is okay, presses the content confirmation button 305c.

[0064] Next, the sentence generation means 104 receives the operation content (S7), generates scout sentence information 114 based on the operation content, stores it in the storage unit 11 (S8), and displays a screen (not shown) for confirming whether to transmit the generated scout sentence information 114 to the user 5a.

[0065] The user 5b checks the content of the screen and operates the terminal 3b to send a transmission request. The terminal 3b sends a transmission request to the information processing device 1 according to the operation content.

[0066] The sentence generation means 104 of the information processing device 1 sends and outputs a scout sentence to the user 5a according to the transmission request (S9). Note that the output is not limited to the transmission of the scout sentence, and it may be the generation of text data. In this case, after the user 5b edits the text data, etc., it is manually transmitted.

[0067] (3) Standard sentence registration and editing operation When the user 5b operates the terminal 3b to manage the standard sentence used in the scout sentence sent when scouting talents (such as pressing the management button 301c' in Fig. 8), a request is sent from the terminal 3b to the information processing device 1, and the following screen is displayed and processed on the display screen of the terminal 3b as a response.

[0068] Fig. 9 is a schematic diagram showing a configuration example of the screen displayed on the display units of the terminals 3a and 3b.

[0069] The screen 30d is a screen displayed when managing the standard sentence information 113, and has an add button 300d for adding a new standard sentence, a standard sentence title 301d of the standard sentence registered as the standard sentence information 113, an edit button 302d for editing the standard sentence, a delete button 303d for deleting the standard sentence, and a close button 304d for ending the management of the standard sentence information 113.

[0070] While checking the content of the screen 30d, the user 5b presses the edit button 302d of the desired standard sentence to edit the content of the standard sentence, and also presses the add button 300d to add a new standard sentence.

[0071] The terminal 3b sends a request to the information processing device 1 according to the operation content, and the following screen is displayed and processed on the display screen of the terminal 3b as a response.

[0072] FIG. 10 is a schematic diagram showing a configuration example of a screen displayed on the display units of terminals 3a and 3b.

[0073] Screen 30e is a screen displayed when editing (adding) fixed-form sentence information 113. It is displayed by pressing the edit button 302d (add button 300d) on screen 30d, and has a fixed-form sentence title input field 300e, a fixed-form sentence content input field 301e, and an update button 302e.

[0074] While user 5b checks the content of screen 30e, the user inputs text or the like into the fixed-form sentence title input field 300e and the fixed-form sentence content input field 301e to edit (add) the content of the fixed-form sentence, and presses the update button 302e when the editing (adding) is completed.

[0075] Terminal 3b sends a request to information processing device 1 according to the operation content. As a response to this, the sentence generation means 104 of information processing device 1 records the input information as fixed-form sentence information 113.

[0076] (Effects of the Embodiment) According to the above-described embodiment, a request is made to extract the user's charm from user information 111 for the generation AI, and a request is made to generate a scout message for the user based on the charm extracted according to the content of the recruitment information 112. Since the generated sentence is used to generate the scout message, it is possible to create a sentence according to the destination user and the purpose of transmission, rather than just combining prepared fixed-form sentences.

[0077] [Other Embodiments] Note that the present invention is not limited to the above-described embodiment, and various modifications are possible without departing from the spirit of the present invention.

[0078] For example, after extracting the user's charm from the user's profile in step S2 and then requesting to understand the content of the recruitment in step S5, the selection of the user to be scouted and the selection of the recruitment may be performed first, and while performing other processes, the operation of extracting the user's charm from the user's profile and the operation of requesting to understand the content of the recruitment may be completed. Usually, since the charm extraction operation and the recruitment content understanding operation in the generation AI2 take time, by proceeding with these operations in parallel with the reception of operations from the user and other processes, the waiting time of the user can be reduced.

[0079] Further, the sentence generation request means 103 may request the generation AI2 to learn from the past sentences of the user of the transmission source and make the generated sentence into a sentence in the style characteristic of the user of the transmission source.

[0080] Further, when requesting the generation of a sentence, the sentence generation request means 103 may request the generation AI2 to touch on the common career of the users who are common friends of the user of the transmission source and the user of the reception destination on the SNS. As a result, since the sentence of the scout includes the talk of the user who is a common friend, it is possible to give the user of the reception destination, who is the reader, a sense of security and trust in the transmission source, and an improvement in the reply rate can be expected.

[0081] Further, the sentence generation request means 103 may request the generation AI2 to learn sentence generation for scout sentences with a high reply rate. As a result, it becomes possible to generate scout sentences with a high reply rate.

[0082] In the above embodiment, the functions of the respective means 100 to 104 of the control unit 10 are realized by a program, but all or part of each means may be realized by hardware such as an ASIC. Also, the program used in the above embodiment can be stored in a recording medium such as a CD-ROM and provided. Also, the replacement, deletion, addition, etc. of the above steps described in the above embodiment are possible within a range that does not change the gist of the present invention.

Explanation of Reference Numerals

[0083] 1: Information processing device 2: Generation AI 3a, 3b: Terminal 4: Network 5a, 5b: User 10: Control unit 11: Memory unit 12: Communication unit 100: User selection means 101: Recruitment selection means 102: Charm extraction request means 103: Article generation request means 104: Article generation means 110: Article generation program 111: User information 112: Recruitment information 113: Fixed-form text information 114: Scout text information

Claims

1. Computer, A user to whom the scout message is to be sent, and a selection means for accepting a selection of previously prepared recruitment information; An extraction request means for requesting a generation AI to extract a plurality of attractions from the user information of the user; A sentence generation program for functioning as a generation request means for understanding the previously prepared recruitment information and requesting the generation AI to generate a sentence for the user based on the extracted multiple attractions.

2. Computer, The text generation program according to claim 1 , further functioning as a generation means for generating a scouting text using the text generated by the generation AI for the user and at least one of a pre-prepared template text.

3. 3. The text generation program according to claim 2, wherein said generation request means requests said generation AI to generate said sentence for a type other than the type of said pre-prepared fixed phrases.

4. 4. The text generation program according to claim 1, wherein the generation request means learns past texts of the sending user and requests the generation AI to convert the texts into the writing style of the sending user.

5. The sentence generation program according to any one of claims 1 to 3, wherein the generation request means, when requesting the generation of the sentence, requests the generation AI to access common carriers of users who are mutual friends of the sending user and the destination user.

6. The text generation program according to any one of claims 1 to 3, wherein the generation request means requests the generation AI to learn sentence generation for scout sentences with high response rates.

7. A user to whom the scout message is to be sent, and a selection means for accepting a selection of previously prepared recruitment information; An extraction request means for requesting a generation AI to extract a plurality of attractions from the user information of the user; An information processing device having a generation request means for understanding the previously prepared recruitment information and requesting the generation AI to generate a sentence for the user based on the extracted multiple attractions.

8. A step of accepting a user to whom the scout message is to be sent and a selection of previously prepared recruitment information; Requesting a generating AI to extract a plurality of attractions from the user information of the user; A sentence generation method comprising a step of requesting the generation AI to understand the pre-prepared recruitment information and generate a sentence for the user based on the extracted multiple attractions.

9. A generation AI that generates at least a sentence in response to a request; An information processing system comprising an information processing device having a user to whom a scouting message is to be sent, a selection means for accepting a selection of pre-prepared recruitment information, an extraction request means for requesting the generation AI to extract multiple attractions from the user information of the user, and a generation request means for requesting the generation AI to understand the pre-prepared recruitment information and generate a message for the user based on the extracted multiple attractions.

Images