Llm based automatic portfolio creation system for job seekers

Abstract

The present invention provides an LLM-based system for automatically generating job seeker portfolios. A system for automatically generating portfolios according to one embodiment may include a career management platform that provides HR-related information, sharing, and individual portfolio management functions; a user DB management server that registers users who access the career management platform and grants specific rights to authorized users; and a data processing server that provides job-specific portfolio automatic generation functions to authorized users who access the career management platform.

Description

Technology Field

[0001] The present invention relates to a technology for a platform and an automatic portfolio generation system that provides convenience features to enhance job seekers' employment capabilities based on an AI model that recognizes and processes natural language. Background Technology

[0002] A Large Language Model (LLM) is a type of deep learning algorithm capable of performing various Natural Language Processing (NLP) tasks. LLMs can be trained on a significant amount of unlabeled text using self-supervised or semi-self-supervised learning.

[0003] LLM primarily uses Transformer models and is trained on vast datasets. The Transformer, first introduced in a paper published by Google, processes sequences in parallel and demonstrates good performance in the field of natural language processing through an attention-based model configuration, unlike models based on RNNs (Recurrent Neural Networks) or CNNs (Convolutional Neural Networks) that process inputs sequentially.

[0004] The Transformer neural network architecture is a very large model containing hundreds of billions of parameters, which allows for the easy collection of vast amounts of data through methods such as crawling. Additionally, data scientists can train Transformer-based LLMs using GPUs, significantly reducing language model training time.

[0005] The basic Transformer is a set of neural networks consisting of encoders and decoders equipped with self-attention capabilities, and the encoders and decoders extract meaning from a series of texts and understand the relationships between words and phrases within the text.

[0006] As LLM is trained on a vast amount of data, it is highly adaptable to flexible environments, and while not perfect, it demonstrates remarkable predictive capabilities based on a relatively small number of prompts or inputs. LLM can be used in generative AI that generates content based on natural language input prompts.

[0007] Recently, various technologies related to generative AI are being developed, but commercialized business models utilizing AI technology are limited, whereas business models applying AI models are considered to have the potential for infinite expansion in their scope of application. The problem to be solved

[0008] According to one embodiment of the present invention, a generative AI technology is presented that automatically and rapidly generates a job seeker-customized portfolio from information input in natural language using LLM as a base technology. means of solving the problem

[0009] According to the first aspect of the present disclosure, a portfolio automatic generation system may include a career management platform that provides HR (human resources) related information, sharing and individual portfolio management functions; a user DB management server that registers users who access the career management platform and grants specific rights to authorized users; and a data processing server that provides job-specific portfolio automatic generation functions to authorized users who access the career management platform.

[0010] The above data processing server provides a function to automatically generate job-specific portfolios using LLM, and the above LLM may be an artificial intelligence language model that is trained based on large-scale text data and has characteristics useful for natural language processing.

[0011] The data processing server described above may include a data collection module that collects history information and activity information for a specific user in conjunction with a user DB; a portfolio creation module that analyzes the context of the collected individual data based on LLM, classifies the meaning of key history, achievements, and activity information according to a predetermined format, and creates a personalized portfolio; and a revision module that compares and analyzes the created personalized portfolio with competencies required by job-related information and target company information, and revises it into a job-tailored portfolio by inserting a predetermined phrase if necessary.

[0012] The data collection module above can collect additional information regarding the specific user based on the tagging information when a user of another account tags the specific user's account.

[0013] The above portfolio creation module may be characterized by extracting history information, including educational background and career experience, centered on a specific user from the information collected by the above data collection module and outputting it as essential data, and, based on the above LLM, identifying the meaning, duration, and role of the activity information related to the specific user based on context to generate natural-sounding phrases introducing the activity information, and creating a personalized portfolio for the specific user in a predetermined format.

[0014] The data collection module may further include receiving feedback on the personalized portfolio created from the specific user through the career management platform, or receiving input from the specific user regarding desired job roles or target companies for employment.

[0015] It may further include a corporate information DB analyzed based on LLM based on the desired talent profile, vision, and job postings of each company, and the revision module may be characterized by retrieving information from the corporate information DB and combining it with portfolio data personalized for the specific user to revise into a job-customized portfolio that includes introductory text reproduced to suit a specific company or specific job.

[0016] The above revision module may include revising into a job-customized portfolio related to the user's desired job, based on an AI model that has learned the optimal portfolio for each job according to NCS. Effects of the invention

[0017] According to one embodiment of the present invention, job seekers can input necessary information into the LLM simply by entering a brief introductory phrase about their activity information, and the implications and job relevance of the input activity information can be analyzed using the LLM to automatically generate an appropriate form and introductory phrase.

[0018] By utilizing LLM to simplify the method of using the platform and providing an automatic portfolio generation function optimally customized for individuals, useful convenience features can be provided to platform users.

[0019] LLM is a generative AI model trained on vast amounts of data that can automatically extract implications from the system, even for aspects the user did not intend at the time of inputting activity information, and present them to the user.

[0020] Multiple versions of portfolios optimized by company and job can be automatically generated based on the job seeker's desired company and job information. Brief explanation of the drawing

[0021] FIG. 1 is a configuration diagram of a portfolio automatic generation system (100) according to one embodiment of the present invention. FIG. 2 is a flowchart regarding distinguishing users with usage rights in a portfolio automatic generation system (100) and providing a job-specific portfolio automatic generation function to users who have undergone a predetermined user verification procedure. FIG. 3 is a flowchart illustrating a workflow for creating a job-specific portfolio in an automatic portfolio creation system (100) and providing it to a user. FIG. 4 is an example of the types of information input through a portfolio automatic generation system (100) according to an embodiment of the present invention. FIG. 5 is an example of job-specific portfolio creation information output through a portfolio automatic creation system (100) according to an embodiment of the present invention. Specific details for implementing the invention

[0022] The embodiments of the present invention described below are provided to more clearly explain the present invention to those skilled in the art, and the scope of the present invention is not limited by the following embodiments, and the following embodiments may be modified in various other forms.

[0023] The terms used herein are for describing specific embodiments and are not intended to limit the invention. Terms used herein in the singular form may include plural forms unless the context clearly indicates otherwise. Additionally, the terms “comprise” and / or “comprising” used herein specify the presence of the mentioned features, steps, numbers, actions, parts, elements, and / or groups thereof, and do not exclude the presence or addition of one or more other features, steps, numbers, actions, parts, elements, and / or groups thereof.

[0024] Furthermore, the term "connection" as used herein refers not only to the direct connection of certain members but also encompasses the concept of indirect connection through the interposition of other members between them. Additionally, when a member is described as being "on" another member in this specification, this includes not only cases where a member is in contact with another member but also cases where another member exists between the two members. The term "and / or" as used in this specification includes any one of the listed items and all combinations of one or more thereof. Furthermore, terms of degree such as "approximately" and "substantially" used in this specification are used to mean a range of numerical values ​​or degrees or approximate values, taking into account inherent manufacturing and material tolerances, and are used to prevent an infringer from unfairly exploiting the disclosures in which precise or absolute figures provided to aid in understanding the invention are mentioned. Additionally, detailed descriptions of known functions and configurations that could obscure the essence of the invention will be omitted in this specification.

[0025] First, the present invention relates to a technology for a platform used in an environment where a network is connected through a communication network. Here, the communication network may be a high-speed backbone network of a large-scale communication network capable of data transmission and reception services, and may be a next-generation wireless network including Wi-Fi, WiGig, Wibro (Wireless Broadband Internet), WiMAX (World Interoperability for Microwave Access), etc., for providing the Internet or high-speed multimedia services.

[0026] The above Internet may refer to a global open computer network structure that provides the TCP / IP protocol and various services existing in the upper layer, namely HTTP (Hyper Text Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), SNMP (Simple Network Management Protocol), NFS (Network File Service), NIS (Network Information Service), etc., and may provide an environment that enables user terminal devices to connect to a server operating the platform.

[0027] Meanwhile, the above-mentioned Internet may be a wired or wireless Internet, and may also be a core network integrated with a wired public network, a wireless mobile communication network, or a mobile Internet.

[0028] If the communication network is a mobile communication network, it may be a synchronous mobile communication network or an asynchronous mobile communication network.

[0029] As an example of the above-mentioned asynchronous mobile communication network, a communication network using the WCDMA (Wideband Code Division Multiple Access) method may be cited. In this case, although not shown in the drawings, the mobile communication network may include, for example, an RNC (Radio Network Controller). Meanwhile, although the above-mentioned WCDMA network was given as an example, it may be a next-generation communication network such as a 3G LTE network, a 4G network, a 5G network, or other IP-based IP networks.

[0030] Next, a management server and a processing server according to one embodiment of the present invention can receive information about a user (e.g., an employed person or a job seeker) from a user's terminal device through a communication network, analyze the received information to provide a portfolio file generated to the user's terminal device, and perform a function of providing guidance such as providing information on matching companies or recommended job positions.

[0031] A user terminal device according to one embodiment of the present invention is a digital device capable of communicating after connecting to a platform, and any digital device equipped with memory means and equipped with a microprocessor to have computational capabilities, such as a desktop computer, a notebook computer, a workstation, a PDA, a web pad, or a mobile phone, can be adopted as a user terminal device according to the present invention.

[0032] In particular, the user terminal device may include a browser-related program that enables the user to receive a user interface provided by the server for inputting user information.

[0033] Below, we will examine the configuration of the portfolio automatic generation system according to the present invention and the functions of each component.

[0034] FIG. 1 is a configuration diagram of a portfolio automatic generation system (100) according to an embodiment of the present invention. Referring to FIG. 1, the portfolio automatic generation system (100) includes a career management platform (10), a user DB management server (110), a data processing server (150), a data collection module (210), a portfolio generation module (220), and a revision module (230). Additionally, the portfolio automatic generation system (100) may include a user DB (50), user information collection (DB), corporate information DB (70), and LLM (Large Language Model, 300).

[0035] The career management platform (10) may be a platform that provides HR (Human Resource) related information, sharing, and individual portfolio management functions. For example, the career management platform (10) may be a platform with user-friendly features implemented to allow platform users to record their history or activity information, edit it, or share it with users of other accounts. Additionally, the career management platform (10) may serve as a community for information sharing for college students or job seekers, and may provide functions for matching job seekers with companies or recommending them.

[0036] In the present invention, the career management platform (10) focuses on explaining the function of providing a personalized job-tailored portfolio to platform users.

[0037] For example, when a personalized portfolio is automatically generated by a data processing server (150), the career management platform (10) can provide it to the user in the form of a file. If the user wishes to modify the personalized portfolio, a UI (User Interface) for this purpose may be provided, or if the user wishes to regenerate the generated personalized portfolio, additional information may be received to provide the user with a revised personalized portfolio.

[0038] To simplify the process of a user creating a personalized portfolio through a career management platform (10), a workflow can be implemented that allows a personalized portfolio to be created automatically with just one or two clicks and a simple phrase input.

[0039] The above career management platform (10) can receive information from the user, provide a portfolio file created by the user, and provide a UI for the process of receiving feedback from the user.

[0040] For example, when the career management platform (10) receives input regarding a user's portfolio theme preference and content that the user wants to be emphasized in the portfolio, the data processing server (150) can search for a portfolio format optimized for the individual and create a personalized portfolio that reflects the individual's preferences.

[0041] As another example, the data processing server (150) can group multiple users based on collected user information such as the user's age, affiliation, desired company, and job information, and extract portfolio topics, job-related terms, and common inclusions characteristic of the group, and generate a recommended portfolio specialized for the group based on this. This is a concept opposite to Big Data, indicating that the AI ​​model can present results more optimized for specific groups based on Small Data.

[0042] The user DB management server (110) can register users who have accessed the career management platform (10) in the user DB (50) and grant an account to users who have gone through a predetermined registration procedure. The account is a unique identification level for recognizing individual users on the platform and mapping various information related to individual users.

[0043] When the connectivity between the user and the account is recorded in the user DB (50), the user DB management server (110) can perform management, such as granting appropriate usage rights or restricting activities to the user or account based on the user DB (50).

[0044] The user DB management server (110) can collect information such as the user's ID and the IP address of the terminal connecting to the platform, and can manage trusted user, terminal, and account information by establishing a predetermined authentication procedure. In addition, the user DB management server (110) can manage users connecting to the platform by restricting the right to use platform functions for unauthorized users or accounts or by granting arbitrary usage rights.

[0045] The data processing server (150) can provide a job-specific portfolio automatic generation function to authorized users who access and are authorized through the career management platform.

[0046] Here, the term "authorized user" was used, but as previously discussed, the service may be provided for specific accounts rather than users, and in the following description, the term "user" may be understood to refer to specific accounts.

[0047] Meanwhile, the aforementioned LLM may be an artificial intelligence language model trained based on large-scale text data and possessing characteristics useful for natural language processing. LLM is a term referring to a type of artificial neural network trained through vast amounts of data, and there are numerous AI models that can be classified as LLMs. In this invention, open-source models with open LLM-related licenses or LLM models that have been improved or modified to suit business models may be used.

[0048] The characteristics of the LLM model used in this invention are that it is an AI model specialized in natural language processing and that it is an AI model trained to generate a portfolio optimized for an individual based on input information. In addition, there may be various other functions unique to the above LLM, and in this invention, the above characteristics have been described merely as examples.

[0049] The data processing server (150) may include a data collection module (210), a portfolio creation module (220), and a revision module (230), and may read data from a user information collection DB (60) and a corporate information DB (70) or store data.

[0050] The user information collection DB (60) may be a DB in which at least one of the basic information, activity information, and desired company / job information is matched and stored for each user registered in the above user DB.

[0051] The data collection module (210) can collect basic information and activity information about a specific user in conjunction with the user DB (50). At this time, the data collection module (210) can classify and store the collected basic information and activity information according to a predetermined DB architecture.

[0052] For example, a data processing server (150) can parse a natural language phrase entered by a user based on LLM, analyze the implications and expression patterns of the natural language phrase, and record the information analyzed by the LLM for each user in a user information collection DB.

[0053] The data collection module (210) may further include receiving feedback on the personalized portfolio created from the specific user through the career management platform, or receiving information regarding the desired job or target company for job seeking from the specific user.

[0054] As another example, the data collection module (210) can collect additional information about the specific user based on the tagging information when a user of another account tags the specific user's account. In this case, the data collection module (210) can store the information collected based on the tagging information in the user information collection DB (60).

[0055] Meanwhile, the data processing server (150) can store or read corporate information analyzed based on LLM in the corporate information DB (70) based on the desired talent profile, vision, and job postings for each company.

[0056] For example, the corporate information DB (70) may be a record of the career and activity details required by the company from job seekers by analyzing past job postings posted by the company. The corporate information DB (70) may contain various types of corporate information, including information regarding the company's industry, business type, industry, and market. At this time, the user's desired job may be classified based on the National Competency Standards (NCS).

[0057] However, in addition to the examples described above, there may be many types of data collected by the data collection module (210), and the data collection module (210) may obtain information about the user through crawling, etc., in addition to the above paths. Alternatively, in cases where consent for the collection / use / utilization of specific personal information is obtained from the individual user, such as in a MyData business, information about the individual may be collected and used according to a predetermined process.

[0058] The portfolio creation module (220) can create a personalized portfolio by analyzing the context of the collected individual data based on LLM and classifying the meaning of key history, performance, and activity information according to a predetermined format.

[0059] For example, if a user writes a one-line phrase introducing their activity content, such as "Participated in the OOO contest at school and won YYY," the portfolio creation module (220) can set the category of the activity content to "contest" based on the above LLM, identify the main content as "won YYY," etc., and extract meaningful content from the activity content.

[0060] Meanwhile, the portfolio creation module (220) can obtain information about the 000 contest or information about the meaning of winning YYY by using the LLM to search the web.

[0061] As another example, the portfolio creation module (220) can extract meaningful content by associating the user's activity content with the desired job.

[0062] Additionally, the portfolio creation module (220) can classify the information obtained by the LLM according to a portfolio format created in a predetermined format and generate a main introductory phrase to be published in the portfolio based on this.

[0063] More specifically, the portfolio creation module (220) may be characterized by extracting history information including educational background and career centered on a specific user from the information collected by the data collection module (210) and outputting it as essential data, and generating natural phrases introducing activity information related to the specific user based on the context of the meaning, activity period, and role of the specific user based on the LLM, and creating a personalized portfolio for the specific user in a predetermined format.

[0064] For example, the portfolio creation module (220) may output history information, including educational background and career history, among the information collected about a specific user, as essential data constituting the portfolio in a designated space within the portfolio form. At this time, the portfolio creation module (220) may sort the educational background and history information in chronological order or design and output the essential data in a timeline format.

[0065] Meanwhile, the portfolio creation module (220) can classify all information collected about a specific user, excluding the essential data among the user information recorded in the user information collection DB (50), as activity information. The portfolio creation module (220) can classify the activity information according to type or period, or classify the activity information according to the method by which the user information was collected.

[0066] According to one embodiment, the portfolio creation module (220) can identify the meaning of the activity information, the duration of the activity, and the role of the specific user based on context using the LLM. For example, the LLM is a language model with a special advantage in natural language processing, and can parse information about activities recorded in the user information collection DB (60) by phrase to distinguish between core information and additional information, or re-evaluate the meaning in the LLM by considering the user's intent when interpreting words with abstract meanings and words expressing emotions.

[0067] Additionally, the portfolio creation module (220) can perform an interpretation of the collected activity information through LLM and generate natural phrases that introduce the activity information. At this time, the portfolio creation module (220) can select an appropriate portfolio form among the pre-learned model portfolio forms and generate natural phrases that introduce the activity information in accordance with the style of the selected portfolio form. Here, the natural phrases may be introductory phrases generated by the portfolio creation module (200) based on the sentence forms, composition, emphasis, etc., mainly used in model portfolios. If the amount of activity information is large, the portfolio creation module (220) can output the collected activity information by placing activities with a predetermined priority set by the user in a position where they can be more visible in the portfolio form.

[0068] As another example, if the collected activity information is significantly more than the amount of information to be output to a predetermined portfolio form, the portfolio creation module (220) may output only the activities having a predetermined priority set by the user to the portfolio form.

[0069] The revision module (230) can compare and analyze the generated personalized portfolio with the competencies required by job-related information and job target company information, and, if necessary, revise it into a job-tailored portfolio by inserting a predetermined phrase.

[0070] The above revision module (230) may be characterized by retrieving information from the above corporate information DB (70), combining it with portfolio data personalized for the above specific user, and revising it into a job-tailored portfolio that includes an introductory phrase reproduced to suit a specific company or specific job.

[0071] Here, the job-specific portfolio may be a portfolio created in a form corresponding to a table of job performance capabilities and basic occupational competencies suitable for document screening or interview screening based on NCS. The revision module (230) may include revising into a job-specific portfolio related to the user's desired job based on an AI model that has learned the optimal portfolio for each job according to NCS.

[0072] Meanwhile, the user may have multiple desired job roles. In this case, the revision module (230) can generate a job-specific portfolio by extracting prompt result values ​​for each specific job role entered by the user. According to this, if the user's desired job role changes, the revision module (230) regenerates the job-specific portfolio, and the previous result values ​​may not affect the revised portfolio.

[0073] As another example, if the user changes the desired company information, the revision module (230) can regenerate the portfolio by reflecting the changed desired company information. At this time, the revision module (230) can revise the portfolio into a job-tailored portfolio by selecting activity information to be emphasized in the changed desired company information, or by changing phrases mainly used in the peer group or a predetermined portfolio format.

[0074] Meanwhile, the user may input information regarding a specific industry group, the industry desired by the user, or the desired job, without specifying and inputting a desired company. At this time, the revision module (230) can interpret the information entered by the user in natural language format based on LLM, identify the user's intent, match the desired company, or revise it into a job-customized portfolio organized around a specific job. Meanwhile, examples of data that can be input into the LLM and examples of output data after passing through the LLM will be further explained later through FIGS. 4 and 5.

[0075] FIG. 2 is a flowchart illustrating how a portfolio automatic generation system (100) distinguishes users with usage rights and provides a job-specific portfolio automatic generation function to users who have undergone a predetermined user verification procedure.

[0076] A portfolio automatic generation system (100) can be implemented according to a predetermined workflow presupposed by a career management platform (10), a user terminal accessing the career management platform, a user (or user ID or account or user's IP) using the user terminal, and a user DB management server (110) and a data processing server (150) that run and operate the platform on a wired or wireless network, by performing a predetermined action according to their respective roles. This is considered as a series of actions to perform a portfolio automatic generation function on the entire system.

[0077] In step S110, the user (or user terminal) can access the career management platform (10).

[0078] In step S120, the user DB management server (110) can determine whether the user who accessed the career management platform (10) is a user registered in the user DB (50).

[0079] In step S130, if the user is not a registered user, the user DB management server (110) presents a screen for the user to input basic information, and if the basic information is entered, it can be registered in the user DB and managed.

[0080] Meanwhile, according to one embodiment, there may be cases where a large amount of user information is entered in bulk through a linked cooperation platform. In this case, the user DB management server (110) can omit the process of presenting a screen for entering basic information to the user, map the large amount of user information according to a predetermined classification standard, register it in the user DB (50), and manage it.

[0081] When the user's basic information input procedure is completed, the user DB management server (110) can respond to a request to check user registration status coming from the career management platform (10) and check whether a specific user is a user registered in the user DB (50).

[0082] In step S140, if it is confirmed whether the user accessing the career management platform (10) is a registered user, the portfolio automatic generation system (100) can provide the user with a job-specific portfolio automatic generation function.

[0083] FIG. 3 is a flowchart illustrating a workflow for creating a job-specific portfolio in an automatic portfolio creation system (100) and providing it to a user.

[0084] In step S210, the data processing server (150) can proceed with subsequent procedures when the user accesses the career management platform (10) and is granted a certain authority.

[0085] In step S220, the data collection module (210) can ask the user whether to enter additional information through the career management platform (10).

[0086] In step S230, if the user wishes to input additional information, the data activity collection module (210) provides the user with a screen for inputting information and receives the input to collect history information and activity information for a specific user. At this time, the data collection module (210) can collect history information and activity information for a specific user in conjunction with the user DB (50).

[0087] In step S240, if a user of another account on the career management platform (10) tags a specific user's account, the data collection module (210) can separately collect history information and activity information for the specific user related to the tagging information.

[0088] Meanwhile, when the above S230 step is completed, the data collection module (210) can store history information and activity information for the specific user in the user information collection DB (60).

[0089] In step S250, the portfolio creation module (220) can create a personalized portfolio by analyzing the context of the collected individual data based on LLM and classifying the meaning of key history, performance, and activity information according to a predetermined format. An example of the creation of a personalized portfolio can be referenced from the content previously described in FIG. 1.

[0090] In step S260, the revision module (230) can compare and analyze the generated personalized portfolio with the competencies required by job-related information and job target company information, and, if necessary, insert a predetermined phrase to revise it into a job-customized portfolio. An example of revising into a job-customized portfolio can be referenced from the content described above through FIG. 1.

[0091] FIG. 4 is an example of the types of information input through a portfolio automatic generation system (100) according to an embodiment of the present invention.

[0092] The embodiment presented through FIG. 4 is merely an example of information input into an LLM, and the process of structuring and categorizing the input information may be configured differently from the above example. This should be understood as material describing an example of the present invention.

[0093] An example of information input into the LLM may include a structured category and data loaded into a lower layer thereof to easily retrieve data from the user information collection DB (60) in the data collection module (210).

[0094] Referring to Fig. 4, an example of information entered into the LLM can be categorized into desired company information (1010), basic information (1010), and activity information (1030). Under each category, information regarding target companies, desired job positions, etc. within the category of desired company information (1010) can be entered. At this time, the desired job positions can be classified based on the National Competency Standards (NCS).

[0095] In addition, information regarding name, education, career, and awards can be entered within the categories of basic information (1020). In addition, information regarding classification, activity name, activity date / period, introduction phrase (Raw), and assigned role can be entered within the categories of activity information (1030).

[0096] FIG. 5 is an example of job-specific portfolio creation information output through a portfolio automatic creation system (100) according to an embodiment of the present invention.

[0097] As with Fig. 4, the output result presented in Fig. 5 is an example that does not take design elements into consideration, and it should be taken into account that the output form can be modified in various ways.

[0098] Referring to FIG. 5, the job-specific portfolio generated by the portfolio automatic generation system (100) can output self-introduction (1510) and activity introduction (1520) information. At this time, the job-specific portfolio may be a portfolio with a predetermined format, introduction method, etc. optimized according to the NCS classification.

[0099] For example, the self-introduction (1510) category may include content regarding self-introduction, slogan, and core competencies, and the activity introduction (1520) category may include content regarding activity name, definition of problem, solution, performance, and job-related insights.

[0100] An embodiment according to the present invention may be implemented in the form of a computer program that can be executed through various components on a computer, and such a computer program may be recorded on a computer-readable medium. In this case, the medium may include a magnetic medium such as a hard disk, a floppy disk, and a magnetic tape, an optical recording medium such as a CD-ROM and a DVD, a magneto-optical medium such as a floptical disk, and a hardware device specifically configured to store and execute program instructions, such as a ROM, RAM, flash memory, etc.

[0101] Meanwhile, the above computer program may be specially designed and configured for the present invention or may be known and available to those skilled in the art of computer software. Examples of computer programs may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0102] According to one embodiment, the method according to various embodiments of the present disclosure may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0103] Unless explicitly stated or contrary to the order of the steps constituting the method according to the present invention, said steps may be performed in a suitable order. The present invention is not necessarily limited by the order in which said steps are described. The use of all examples or exemplary terms (e.g., etc.) in the present invention is merely for the purpose of describing the present invention in detail, and the scope of the present invention is not limited by said examples or exemplary terms unless limited by the claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and changes may be made according to design conditions and factors within the scope of the claims or equivalents to which they are added.

[0104] The scope of the present invention is not limited to the embodiments described above, and all scopes equivalent to or equivalently modified from the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention. Explanation of the symbols

[0105] 10: Career Management Platform 50: User DB 60: User Information Collection DB 70: Corporate Information DB 100: Automated Portfolio Generation System 110: User DB Management Server 150: Data processing server 210: Data Collection Module 220: Portfolio Creation Module 230: Revision Module 300: Large Language Model (LLM)

Claims

Claim 1 A career management platform that provides HR-related information, sharing, and individual portfolio management functions; a user DB management server that registers users accessing the career management platform and grants specific permissions to authorized users; and a data processing server that provides a job-customized portfolio automatic generation function to authorized users accessing the career management platform; wherein the data processing server comprises: a data collection module that collects history information and activity information regarding a specific user in conjunction with the user DB; a portfolio generation module that generates a personalized portfolio by analyzing the context of the collected individual data based on LLM and classifying the meaning of key history, performance, and activity information according to a specific format; and a revision module that compares and analyzes the generated personalized portfolio with competencies required by job-related information and target company information, and revises it into a job-customized portfolio by inserting specific phrases if necessary.The method includes, wherein the data collection module collects additional information regarding the specific user based on the tagging information when a user of another account tags the specific user's account, and the revision module regenerates the portfolio by reflecting the changed desired company information when the user changes the desired company information, and is characterized by selecting activity information to be emphasized in the changed desired company information or revising it into a job-tailored portfolio by changing the wording or portfolio format used in the peer group, wherein the tagging information may be evaluated as a reliability or priority value based on at least one of the relationship between the tagging subject and the user, the tagging frequency, or the tagging time, and the evaluation result may be reflected when configuring the portfolio, and the portfolio creation module extracts history information including educational background and career centered on the specific user from the information collected by the data collection module and outputs it as essential data, and generates natural wording introducing activity information related to the specific user by identifying the meaning of the activity information, the activity period, and the role of the specific user based on the context using the LLM, and a personalized for the specific user in a predetermined format An automatic portfolio generation system characterized by generating a portfolio. Claim 2 In claim 1, the data processing server provides a function to automatically generate a job-tailored portfolio using an LLM, and the LLM is an artificial intelligence language model that is trained based on large-scale text data and is a language model having characteristics useful for natural language processing, a portfolio automatic generation system. Claim 3 delete Claim 4 delete Claim 5 A portfolio automatic generation system according to claim 1, wherein the portfolio generation module extracts history information including educational background and career centered on a specific user from the information collected by the data collection module and outputs it as essential data, and generates natural phrases introducing activity information related to the specific user by identifying the meaning of the activity information, the activity period, and the role of the specific user based on the context using the LLM, and generates a personalized portfolio for the specific user in a predetermined format. Claim 6 delete Claim 7 delete Claim 8 A portfolio automatic generation system according to claim 1, wherein the revision module includes revising into a job-customized portfolio related to the user's desired job based on an AI model that has learned an optimal portfolio for each job according to NCS.

Images