A conversation-based personalized career description creation system and its operation method

Abstract

A method for generating a user-customized career description is disclosed. A method of operation of a user-customized career description generation system according to one aspect of the technical concept of the present disclosure comprises: collecting metadata regarding a user from a user; generating at least one query related to the user; generating additional queries regarding the user using a first artificial intelligence model that takes the at least one query and a user response thereto as input; evaluating the completeness of user information using a second artificial intelligence model that takes the metadata and the query response with the user as input; and generating a career description using a third artificial intelligence model that takes the collected metadata and the query response with the user as input when the completeness of the user information is greater than or equal to a preset standard.

Description

Technology Field

[0001] The technical concept of the present disclosure relates to a conversation-based user-customized career description generation system and a method of operation thereof. More specifically, it relates to a method of providing a service that dynamically identifies a user's career through conversational interaction and provides a career description in real-time in an optimal format for each individual. Background Technology

[0003] Due to the digitalization of information and the rapid advancement of artificial intelligence, the importance of user-customized information delivery technology is steadily increasing across various industries. Particularly in fields such as recruitment, education, counseling, and customer service, where the characteristics and context of individual users are directly linked to the efficiency of information delivery, the recipient's understanding and utilization outcomes can vary significantly depending on how the same information is structured and presented, even if it is the same piece of information.

[0004] Specifically, while job seekers must effectively document their careers and competencies during the application process, they often struggle to structure and systematize this information or to write it specifically in accordance with the requirements of the job posting. Due to these limitations, key information needed by recruiters is omitted, or the wording is overly vague, leading to a decline in the completeness and usability of the documents.

[0005] Meanwhile, recent advancements in Natural Language Processing (NLP) and Large Language Model (LLM) technologies are presenting new possibilities to resolve these issues, making it possible to automatically generate user-customized career summaries beyond existing simple text generation or summarization functions. The problem to be solved

[0007] Existing text generation and summarization technologies primarily focus on static, rule-based processing and often remain at the level of simply shortening text length, extracting key keywords, or modifying formatting. Because this approach fails to adequately reflect individual users' actual career backgrounds or the requirements of the job they are applying for, it results in generated documents that do not match the information recruiters need or have low utility. Furthermore, existing personalized document creation assistance systems are limited to providing specific templates or auto-filling fields, which restricts their ability to supplement incomplete parts of documents and create higher-quality content.

[0008] Accordingly, the present disclosure is devised to overcome the limitations of the aforementioned background technology and existing simple document creation automation technology, and relates to an apparatus and method capable of automatically generating a user-customized career description by analyzing responses through real-time conversation with a user, dynamically generating additional queries, and evaluating the specificity of the responses and their suitability with the job posting.

[0009] However, the problems to be solved in this disclosure are not limited to those mentioned above, and other unmentioned problems may be clearly understood based on the description below. means of solving the problem

[0011] A method for generating a user-customized career description according to one embodiment of the present disclosure is disclosed for realizing the aforementioned objectives. A method of operation of a user-customized career description generation system according to one aspect of the technical concept of the present disclosure comprises: collecting metadata regarding a user from a user; generating at least one query related to the user; generating additional queries regarding the user using a first artificial intelligence model that takes the at least one query and the user's response thereto as inputs; evaluating the completeness of user information using a second artificial intelligence model that takes the metadata and the query-response with the user as inputs; and generating a career description using a third artificial intelligence model that takes the collected metadata and the query-response with the user as inputs when the completeness of the user information is greater than or equal to a preset standard.

[0012] A user-customized career description generation device according to an embodiment of the present disclosure is disclosed for realizing the aforementioned objectives. The device comprises a processor including at least one core, a memory including program codes executable on the processor, metadata regarding a user, a network unit for receiving a response from the user, and a user interface for displaying a generated career description. The processor is configured to collect metadata regarding a user from the user, generate at least one query related to the user, generate additional queries regarding the user using a first artificial intelligence model that takes the at least one query and the user's response thereto as input, evaluate the completeness of user information using a second artificial intelligence model that takes the metadata and the query-response with the user as input, and if the completeness of the user information is above a preset standard, generate a career description using a third artificial intelligence model that takes the collected metadata and the query-response with the user as input, and provide the career description to the user through the user interface. Effects of the invention

[0014] According to the method for generating a user-customized career description based on the technical concept of the present disclosure, the limitations of simple static template-based document creation can be overcome, and the completeness of the career description can be systematically improved by analyzing user responses through real-time conversation-based interaction.

[0015] According to the method for generating a user-customized career description of the technical concept of the present disclosure, by utilizing natural language processing and machine learning technologies to precisely extract key elements from a user's response and dynamically generating follow-up questions based thereon, missing information can be supplemented and the specificity of the document can be enhanced.

[0016] According to the method for generating a user-customized career description of the technical concept of the present disclosure, by quantitatively evaluating the level of specificity of the response and appropriately performing drill-down, expansion, and validation questions based on the result, a career description that can reveal the user's experience and capabilities in multiple layers can be created.

[0017] According to the method for generating a user-customized career description of the technical concept of the present disclosure, a career description optimized for a specific job can be automatically generated by utilizing a job description as input to evaluate the degree of matching with the response, suggesting supplementary questions for missing items, and prioritizing experiences with a high degree of matching.

[0018] According to the method for generating a user-customized career description of the technical concept of the present disclosure, the system can learn the user's expression habits and preferences through user feedback and a process of repeated responses, thereby enabling the creation of increasingly sophisticated customized documents and ensuring the adaptability of the system.

[0019] In particular, the user-customized career description generation system of the technical concept of the present disclosure appears to be highly effective for job seekers in their 40s and 50s who are preparing for re-employment. That is, job seekers who have extensive experience but struggle to systematically reorganize it to fit the latest hiring environment and required competencies can substantially strengthen their competitiveness for re-employment by structuring their experience and reorganizing it into a form optimized for job postings through the device according to the present disclosure.

[0020] The method for generating a user-customized career description based on the technical concept of the present disclosure can be applied to various fields such as recruitment, education, counseling, and writing personal statements. In particular, it can resolve the information gap caused by a lack of document writing skills and maximize the efficiency and usability of information delivery by efficiently providing essential information to recruiters. Brief explanation of the drawing

[0022] FIG. 1 is a conceptual diagram of a user-customized career description generation system according to one embodiment of the present disclosure. FIG. 2 is a block diagram of a user-customized career description generation device according to one embodiment of the present disclosure. FIG. 3 is a block diagram of a user-customized career description generation device according to one embodiment of the present disclosure. FIG. 4 is an exemplary diagram illustrating a literacy dynamic reasoning algorithm and a user interface according to one embodiment of the present disclosure. FIGS. 5 and 6 are exemplary diagrams illustrating the flow of a method for generating a user-customized career description according to one embodiment of the present disclosure. FIG. 7 is an exemplary diagram showing a user-customized career description generated according to the present disclosure. FIG. 8 is a flowchart illustrating the operation method of a user-customized career description generation device according to one embodiment of the present disclosure. Specific details for implementing the invention

[0023] Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed herein are provided merely for the purpose of explaining embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention may be implemented in various forms and are not limited to the embodiments described herein.

[0024] Embodiments according to the concept of the present invention may be subject to various modifications and may take various forms; therefore, embodiments are illustrated in the drawings and described in detail in this specification. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes modifications, equivalents, or substitutions that fall within the spirit and scope of the present invention.

[0025] Terms such as "first" or "second" may be used to describe various components, but said components shall not be limited by said terms. For the sole purpose of distinguishing one component from another, for example, without departing from the scope of rights according to the concept of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component.

[0026] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. Expressions describing the relationships between components, such as "between," "exactly between," or "directly adjacent to," should be interpreted in the same way.

[0027] The terms used herein are used merely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to specify the existence of the described features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0028] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this specification.

[0030] In this specification, the term "processor" may refer to hardware capable of performing functions and operations according to each name described in this specification, computer program code capable of performing specific functions and operations, or an electronic recording medium loaded with computer program code capable of performing specific functions and operations.

[0031] In other words, the term "processor" may refer to a functional and / or structural combination of hardware for carrying out the technical concept of the present invention and / or software for driving said hardware.

[0032] The first to third artificial intelligence models of the present invention may be the same artificial intelligence model, or they may be models individually tuned to be optimized for different tasks. Additionally, if necessary, the first to third artificial intelligence models of the present invention may be different artificial intelligence models.

[0033] The method for generating a user-customized career description according to the present invention can be performed based on the STAR (Situation-Task-Action-Result) framework according to the Context-Driven Sequential Inquiry (CSI)-based information extraction method. The STAR framework is a structural technique that encourages users to describe their experiences or cases by classifying them into four elements: Situation, Task, Action, and Result. In the present invention, by utilizing this framework, responses provided by users can be collected and refined more systematically.

[0034] First, context fidelity is measured for each element of Situation, Task, Action, and Result. Context fidelity refers to a metric that quantifies the sufficiency and clarity of the information provided by the user in the response regarding a specific element. For example, if the description of the Situation is ambiguous or insufficient, the fidelity of that element may be evaluated as low.

[0035] When elements with low fidelity are identified, the system generates a targeted question regarding them. This is a step that elicits additional explanation from the user by presenting questions directly aimed at that area to compensate for the deficiencies.

[0036] In addition, the system detects implicit signals included in the user's response. Implicit signals refer to potential clues that are not explicitly mentioned by the user but are revealed through vocabulary choice, emotional tone, indirect expressions, etc.

[0037] Detected implicit signals can be extended into latent information, and based on this, the system infers contextual information that the user has not directly mentioned. For example, latent information regarding the difficulty of project schedule management can be derived from the response "I was short on time."

[0038] For this inferred potential information, a validation question is generated to verify whether the user has actually experienced it. Through the validation question, the system can directly confirm the reliability of the inferred information from the user.

[0039] Finally, if the user responds positively to the verification question, the system converts that information into a quantification question. This question encourages the expression of inferred experience or performance using numerical and objective indicators, and can be utilized as specific and comparable data when writing a career description.

[0040] Therefore, the present invention is characterized by the ability to multi-layeredly extract and refine user experience through a series of processes based on the STAR framework, including fidelity measurement → targeting refinement → implicit signal detection → latent information inference → verification → quantification. The career description generated through the present invention has increased systematicity and reliability, and can effectively improve the degree of matching with job postings.

[0042] Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, the scope of the patent application is not limited or restricted by these embodiments. Identical reference numerals in each drawing indicate identical components.

[0044] The present invention relates to an apparatus and method for generating a user-customized career description, and in particular includes a Context-based Sequential Inquiry (CSI) mechanism that analyzes a user's response and dynamically generates optimized follow-up questions.

[0045] FIG. 1 is a conceptual diagram of a user-customized career description generation system according to one embodiment of the present disclosure.

[0046] Referring to FIG. 1, a user-customized career description generation system (1000) according to one embodiment of the present disclosure includes a user device (100), a user interface providing device (200) (hereinafter, user-customized career description generation device (200)), and / or a database (20). According to an embodiment, the user-customized career description generation system (1000) may include a plurality of users.

[0047] A user-customized career description generation device (200) according to one embodiment of the present disclosure may receive metadata regarding the user and / or question-and-answer data with the user from the user through a user device (100). The career description generation device (200) may ask the user a question through a text-based conversation and receive response information thereto, and may ask the user a question through a voice-based conversation and obtain response information thereto.

[0048] For example, metadata received through a user device (100) may include the user's personal information and / or information regarding the user's career and work. The user's personal information received from the user through the user device (100) may refer to structured and general attribute values ​​distinct from information regarding career and work, and such data may be used as basic material to supplementarily explain information regarding the user's career and work, or to structure information regarding career and work further to create a document. Meanwhile, information regarding career and work refers to specific information regarding job activities and work experience actually performed by the user.

[0049] More specifically, 'user's personal information' within 'metadata' may include information regarding the user's age, gender, educational background (school name, major, degree, year of graduation, highest level of education), nationality and place of residence, language proficiency, place of residence, place of work, certifications, licenses, awards, desired job, and desired industry sector. 'User's career and work information' within 'metadata' may include the name of the company and organization where the user worked, rank / position (e.g., junior, senior, team leader, manager, etc.), period of employment (hire date, resignation date), major industry sectors (IT, finance, manufacturing, medical, education, public sector, etc.), job function (e.g., development, sales, planning, etc.), assigned duties and roles, projects performed and scope of participation, achievements (quantitative or qualitative performance indicators), technologies and tools used (programming languages, software, equipment, etc.), team size and collaboration type, and assigned rank or position; it may also include information on certifications held, technical skills, and past application history (when applied to a recruitment system). At this time, some metadata can be received as text, and some data can be received as voice and then converted into text and stored. The system according to the present invention can be utilized to systematically reflect the user's background and career by comprehensively collecting and analyzing metadata to ask real-time questions and to automatically generate a career description including the answers to the questions.

[0050] A user-customized career description generation device (200) according to one embodiment of the present disclosure can generate one or more queries related to a user based on collected metadata and transmit them to a user device (100) through a network interface. The user can input an answer to the query through a user interface (UI) of the user device (100), and the user-customized career description generation device (200) can receive the input response and perform subsequent processing.

[0051] According to another embodiment, the user-customized career description generation device (200) of the present disclosure may provide general and fixed questions required for writing a career description to a user device (100) regardless of metadata, and the user may input a response to this into the user device (100). That is, a predefined basic query (common query) may be first transmitted to the user device (100), and then additional customized queries may be sequentially generated and provided according to the user's response result.

[0052] A user-customized career description generation device (200) according to one embodiment of the present disclosure can generate additional queries regarding a user by using a first artificial intelligence model that takes at least one query and a user response to at least one query as input. A user-customized career description generation device (200) according to another embodiment can generate additional queries regarding a user by using metadata regarding a user, at least one query, and a user response to at least one query as input. The 'additional queries' may be user-customized queries necessary for creating a user career description.

[0053] A user-customized career description generation device (200) can generate a context vector by analyzing at least one query and a user response thereto through the first artificial intelligence model. The user-customized career description generation device (200) can extract key elements from the context vector, calculate a question generation probability distribution based thereon, and then select the question with the highest expected value of information acquisition from the question generation probability distribution as an additional query. The question generation probability distribution refers to a distribution in which the selection probability for each of the various types of questions that the system can generate is calculated based on key elements (e.g., keywords, emotional tone, depth of information, etc.) extracted from the context vector.

[0054] According to one embodiment, the 'question generation probability distribution' may be a probability distribution of drill-down, expansion, and validation. 'Drill-down' is a question that further limits the content mentioned in the user's response or requires specific details; 'expansion' is a question that allows exploration of surrounding circumstances or background information related to the response; and 'validation' is a question intended to explicitly confirm information inferred by the system or content indirectly implied by the user, or to validate specific content.

[0055] 'Expected Information Gain' is a value calculated in the form of an expectation representing the actual amount of information that can be obtained from the user by each question type. For instance, if the depth of information in the user's existing responses is shallow and specific keywords are not derived based on key elements extracted from the context vector, the expected information gain may be highest because the diversity and depth of responses obtained through detailed inquiry are significant. Conversely, if keywords possessing a certain level of depth and specificity are extracted from the user's existing responses based on key elements (e.g., keywords, emotional tone, depth of information) derived from the context vector, the expected information gain may appear high due to the potential to secure additional information, as responses obtained through extended inquiry expand the conversation to new topics. If necessary, when some of the user's existing responses are contradictory based on key elements (e.g., keywords, emotional tone, depth of information) derived from the context vector, the expected information gain for verification inquiry may be the highest. Therefore, the system determines the subsequent question by considering both the question generation probability distribution and the expected information gain.

[0056] According to one embodiment, the probability distribution of question generation may be Drill-down Question: 70%, Expansion Question: 20%, and Validation Question: 10%.

[0057] In this way, when an expected value of information acquisition is assigned to each question type, the question with the highest expected value of information acquisition in the question generation probability distribution can be selected as an additional query.

[0058] According to another embodiment, the question generation probability distribution can calculate the probability distribution for each specific question. For example, if a specific user mentions "experience in data analysis," the system can assign probabilities such as 0.6 to the question "What language do you use?", 0.3 to the question "What was the project duration?", and 0.1 to the question "How many people participated?". In this way, the probability distribution is a numerical value representing which of the multiple candidate questions is most closely related to the current context. For example, the question "What language do you use?" may yield a high expected value for information acquisition because various responses such as Python, Java, and C++ are expected, whereas, conversely, information already collected as metadata, such as "What is your name?", may yield a low expected value for information acquisition due to the low diversity of responses.

[0059] 'Key elements' can refer to elements of a group including key keywords, emotional tone, and depth of information.

[0060] The user-customized career description generation device (200) can generate additional queries. The user-customized career description generation device (200) can construct an answer branching tree and calculate information entropy for the branches. The user-customized career description generation device (200) can apply personalized weights based on the user's past response patterns and calculate a priority score for the question by combining the entropy and the personalized weights. The user-customized career description generation device (200) can generate additional queries regarding the user by selecting the question with the highest priority score as an additional query, while applying question diversity constraints to prevent the repetition of the same or similar questions.

[0061] An 'Answer Branching Tree' refers to a tree-shaped data structure that hierarchically organizes candidate follow-up questions that can be derived based on a user's response. 'Information Entropy' is a value that quantifies the degree of uncertainty in responses a specific question can obtain from a user; a higher value indicates a greater potential amount of information to be secured. 'Personalization Weight' is an adjustment coefficient assigned to specific question types that reflects a user's past response patterns or characteristics, enabling the generation of personalized queries. Finally, a 'Diversity Constraint' refers to a constraint that limits the repetition of identical or similar questions to prevent questions from being biased toward specific categories or modes of expression.

[0062] The user-customized career description generation device (200) may include a step of updating a context vector using additional responses from the user to additional queries. This is intended to form a more sophisticated and consistent user profile by reinforcing the semantic space of the context vector by reflecting the response whenever a new response is provided by the user.

[0063] For example, if the user briefly describes their job experience in the initial response, detailed information such as the specific project name, role performed, skills used, and performance indicators of that experience can be supplemented through an additional question and answer process. At this time, the device (200) accumulates and updates these detailed items in the context vector to progressively express the user's job competencies and career characteristics in high resolution.

[0064] In addition, meta-elements such as emotional tone or the depth of information in the response can also be subject to updates. If a user's response includes positive or confident expressions, or contains uncertain vocabulary, these characteristics can be reflected in the context vector and incorporated during the subsequent question generation process or the creation of the final document.

[0065] Therefore, this step goes beyond simply collecting user responses and plays a crucial role in progressively updating knowledge representations through interactive communication, ultimately enabling the automatic generation of career descriptions optimized for individual users.

[0066] The user-customized career description generation device (200) can evaluate the level of specificity of the user's response to additional questions. Specificity refers to the degree to which the response provided by the user remains at a general and comprehensive level, or includes objectively verifiable details such as specific experiences, roles, periods, and performance indicators. For example, if the user simply responds that they "performed a project," it may be evaluated as a low level of specificity, and if it includes a time period, role, and performance, such as "participated as a team leader for the ○○ project from May 2023 to October 2023 and achieved a 20% increase in sales," it may be evaluated as a high level of specificity. If necessary, specificity may be expressed on a scale from 0 to 10. If the level of specificity is below a threshold, the user-customized career description generation device (200) can generate a detailed extraction question that further limits the content mentioned in the response or requires details, and if the level of specificity is above a threshold, it can generate an expanded exploration question that allows for the exploration of surrounding circumstances or background information related to the response.

[0067] The user-customized career description generation device (200) can evaluate the completeness of user information by using a second artificial intelligence model that takes metadata and question-and-answer with the user as input.

[0068] A user-customized career description generation device (200) can generate a career description based on collected metadata and question-and-answer with the user when the completeness of user information is above a preset standard.

[0069] The user-customized career description generation device (200) can repeat the additional query generation step and the completeness evaluation step when the user information completeness is below a preset standard.

[0070] A user-customized career description generation device (200) can collect job descriptions and generate career descriptions corresponding thereto. The user-customized career description generation device (200) can generate additional queries regarding the user by using a first artificial intelligence model that takes at least one query, a user response to the at least one query, and the job description as input. Additionally, the user-customized career description generation device (200) can evaluate the degree of matching between the job description and the response to the additional query by using a second artificial intelligence model that takes the job description, metadata, and the history of questions and answers with the user as input. At this time, if the degree of matching is below a preset standard, supplementary questions are generated for the unmatched requirements, and if the degree of matching is above the preset standard, a career description can be generated by arranging experiences in order of high suitability with the job description. If necessary, the degree of matching evaluation may be performed based on at least one of the required competencies, required years of experience, technical keywords, and job suitability indicators listed in the job description.

[0071] 'Matching' is an indicator representing the extent to which requirements presented in a job posting are met through user responses and experiences. In other words, it is a quantitative evaluation of the degree to which specific requirements are satisfied by a user's career description or responses.

[0072] 'Relevance' is an indicator representing how effectively each user's experience appeals to the requirements of a job posting. It is used as a criterion to identify experiences that better align with the purpose and context of the job posting, even among those that meet the same requirements, and to prioritize their placement.

[0073] According to the present disclosure, a user-customized career description generation device can automatically generate a career description based on the user's metadata and the results of a question and answer with the user, and can also optionally collect job description information to create a career description with a high degree of matching with the competencies and tasks required in the job description.

[0074] FIG. 2 is a block diagram of a user-customized career description generation device according to one embodiment of the present disclosure.

[0075] A user-customized career description generation device (200) according to one embodiment of the present disclosure may be a hardware device or part of a hardware device that performs comprehensive processing and computation of data, or it may be a software-based computing environment connected to a communication network. For example, the user-customized career description generation device (200) may be a server that performs intensive data processing functions and is an entity that shares resources, or it may be a client that shares resources through interaction with the server. Additionally, the user-customized career description generation device (200) may be a cloud system that enables multiple servers and clients to interact to comprehensively process data. Since the above description is merely one example regarding the type of user-customized career description generation device (200), the type of user-customized career description generation device (200) may be configured in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

[0076] Referring to FIG. 2, a user-customized career description generation device (200) according to one embodiment of the present disclosure may include a processor (210), a memory (220), and / or a network unit (230). However, since FIG. 2 is merely an example, the user-customized career description generation device (200) may include other components for implementing a computing environment. Additionally, only some of the disclosed components may be included in the user-customized career description generation device (200).

[0077] A processor (210) according to one embodiment of the present disclosure may be understood as a constituent unit comprising hardware and / or software for performing computing operations. For example, the processor (210) may read a computer program and perform data processing for machine learning. The processor (210) may perform user literacy data generation or customized text generation by executing a first artificial intelligence model and / or a second artificial intelligence model stored in memory (220). An artificial neural network (or neural network) model that may be implemented by the processor (210) may include a statistical learning algorithm that mimics biological neurons in machine learning and cognitive science. According to one embodiment, an artificial intelligence model that may be implemented by the processor (210) according to an embodiment of the present disclosure may include a natural language processing model and / or a natural language generation model. An artificial intelligence model according to one embodiment of the present disclosure may be a pre-trained natural language processing model and / or natural language generation model. The processor (210) can process computational processes such as processing input data for machine learning, extracting features for machine learning, and calculating errors based on backpropagation.

[0078] A processor (210) for performing such data processing may include a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), a tensor processing unit (TPU), an application specific integrated circuit (ASIC), or a field programmable gate array (FPGA). Since the above-described type of processor (210) is merely an example, the type of processor (210) may be configured in various ways within a range understandable to those skilled in the art based on the contents of this disclosure.

[0079] A processor (210) according to one embodiment of the present disclosure can control a series of processes for generating a user-customized career description using metadata received from a user, a user response to a query, etc. The specific operation details of the processor (210) are described later in FIG. 3.

[0080] A memory (220) according to one embodiment of the present disclosure may be understood as a configuration unit comprising hardware and / or software for storing and managing data processed by a user-customized career description generation device (200). That is, the memory (220) may store data of any form generated or determined by the processor (210) and data of any form received by the network unit (230). For example, the memory (220) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory, RAM (random access memory), SRAM (static random access memory), ROM (read-only memory), EEPROM (electrically erasable programmable read-only memory), PROM (programmable read-only memory), magnetic memory, a magnetic disk, or an optical disk. Additionally, the memory (220) may include a database system that controls and manages data in a predetermined system. Since the above-described type of memory (220) is merely an example, the type of memory (220) can be configured in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

[0081] A network unit (230) according to one embodiment of the present disclosure may be understood as a configuration unit that transmits and receives data through any known form of wired or wireless communication system. For example, the network unit (230) may perform data transmission and reception using wired or wireless communication systems such as a local area network (LAN), wideband code division multiple access (WCDMA), long term evolution (LTE), wireless broadband internet (WiBro), 5th generation mobile communication (5G), ultrawide-band wireless communication, ZigBee, radio frequency (RF) communication, wireless LAN, wireless fidelity (Wi-Fi), near field communication (NFC), or Bluetooth. Since the communication systems described above are merely examples, wired or wireless communication systems for data transmission and reception of the network unit (230) may be applied in various ways other than those described above.

[0082] According to one embodiment, the network unit (230) may receive metadata regarding a user, original text for which the user requests analysis, a user response to a query, and / or user feedback from the user device. The network unit (230) may transmit at least one query and / or custom text to the user device.

[0084] FIG. 3 is a block diagram of a user-customized career description generation device according to one embodiment of the present disclosure. FIG. 4 is an example diagram showing an additional query generation algorithm, a response evaluation algorithm, and a user interface according to one embodiment of the present disclosure.

[0085] Referring to FIG. 3, a user-customized career description generation device (300) according to one embodiment of the present disclosure may include a data collection module (310), a query generation module (320), a response evaluation module (330), a document generation module (340), and a user interface module (350).

[0086] The data collection module (310) can collect and store user metadata (e.g., name, age, gender, educational background, major, job field, certifications, desired job, company name, period of employment, job field, assigned role, project history, achievements, skills used, etc.). This data may be entered directly by the user through the user device (100), or may be automatically collected by linking with an existing history management system, a recruitment platform, or an external database. Referring to FIG. 4, the data collection module (310) transmits the collected data to the query generation module (320).

[0087] Referring to FIG. 3, the query generation module (320) generates an initial query suitable for the user's background based on collected metadata. When the query generation module (320) receives a response to a query related to the user, it can generate an additional query related to the user by executing a first artificial intelligence model based on at least one query and the user's response thereto. The query generation module (320) can generate a context vector by analysis, extract key elements from the context vector, calculate a question generation probability distribution based thereon, and select the question with the highest expected value of information acquisition from the question generation probability distribution as an additional query.

[0088] The query generation module (320) can update the context vector using the user's additional response to the additional query.

[0089] The response evaluation module (330) can evaluate the specificity level of the user response to an additional query. As a result of the evaluation, if the specificity level is below a threshold, the query generation module (320) can generate a detailed inquiry, and if the specificity level is above the threshold, the query generation module (320) can generate an extended inquiry.

[0090] FIGS. 5 and 6 are exemplary diagrams illustrating the flow of a method for generating a user-customized career description according to one embodiment of the present disclosure.

[0091] Referring to FIGS. 5 and 6, when a user inputs a response to an additional query, the response evaluation module (330) calculates the specificity level of the response on a scale of 0 to 10. For example, if the user answers, “I proceeded with the development of an e-commerce platform,” the specificity level may be evaluated as 3 / 10, and it may be determined that detailed information such as project scale, role, and technology is insufficient. In this case, the query generation module (320) can generate a drill-down question and present a supplementary question to the user's device, such as “What was the daily transaction volume of the platform?” If the user answers, “It was about 1 billion won per day on average,” the specificity level is adjusted upward, and then the query generation module (320) generates an expansion question, “What technical challenges were there to handle that scale?” If the user answers, “It was difficult to handle concurrent connections,” a question may be presented to check potential performance again, “How much has performance improved after resolving the concurrent connection issue?” If the user answers, “Response speed improved by 70%, server costs reduced by 40%,” the user’s performance data can be transmitted to the document generation module (340).

[0093] This process can be summarized as shown in [Table 1].

[0094] N0 Question types Example question User response Information Q1 Basic questions What projects did you work on? I developed an e-commerce platform. Derivation of core themes Q2 Detailed inquiry What was the daily transaction volume of the platform? 1 billion won per day Project scale Q3 extended inquiry What technical challenges were there in handling that scale? Concurrent connection processing problem Technical challenges Q4 Performance inference (verification) How much did performance improve after resolving the issue? 70% improvement in response speed, 40% reduction in server costs Performance indicators

[0096] The response evaluation module (330) can evaluate the completeness of user information using a second artificial intelligence model with metadata and question-and-answer with the user as input. If the completeness of user information is above a preset standard, the document generation module can generate a career description based on the collected metadata and question-and-answer with the user.

[0098] Referring again to FIG. 3, the response evaluation module (330) analyzes metadata and user responses at each stage to calculate the completeness of the information. If the completeness is below a preset standard, it collaborates with the query generation module (320) to present a new detailed inquiry or an extended inquiry, and if the completeness is above the standard, it proceeds to the document generation stage.

[0099] Referring to FIG. 4, the document generation module (440) generates a user-customized career description by combining metadata transmitted from the data collection module (410) and in-depth response data collected through the response evaluation DB query generation module (420). At this time, the document generation module (440) can structure and describe the project description, performance figures, skills used, roles, etc., by item.

[0100] The user interface module (460) provides the document generation results to the user device and allows the user to input feedback regarding the suitability, accuracy, and presentation method of the document. The user's feedback is reflected in the document generation module (440) and the query generation module (420), so that additional supplementary questions may be generated or existing documents may be modified if necessary.

[0101] Accordingly, the user-customized career description generation device according to the present embodiment can automatically generate a career description of progressively high quality by providing an initial query based on metadata and career data, and dynamically repeating the generation and analysis of detailed inquiries and extended inquiries according to the user's response.

[0103] FIG. 7 is an exemplary diagram showing a user-customized career description generated according to the present disclosure.

[0104] Referring to FIG. 7, the user interface module (350) provides a question / answer-based interface, outputs the generated career description to the user, and can collect user feedback.

[0106] FIG. 8 is a flowchart illustrating the operation method of a user-customized career description generation device according to one embodiment of the present disclosure.

[0107] A user-customized career description generation device according to one embodiment of the present disclosure can collect metadata about the user (S600).

[0108] A user-customized career description generation device according to one embodiment of the present disclosure can generate metadata or at least one query related to the user (S602).

[0109] A user-customized career description generation device according to one embodiment of the present disclosure can generate additional queries regarding a user by using a first artificial intelligence model that takes at least one query and a user response to at least one query as input (S604). The first artificial intelligence model may include a natural language processing model and a machine learning model.

[0110] The operation of generating additional queries regarding a user of a user-customized career description generation device may include the operation of generating a context vector by analyzing at least one query and a user response thereto through the first artificial intelligence model, the operation of extracting key elements from the context vector and calculating a question generation probability distribution based thereon, and the operation of selecting the question with the highest expected value of information acquisition from the question generation probability distribution as an additional query.

[0112] A user-customized career description generation device according to one embodiment of the present disclosure can evaluate the completeness of user information by using a second artificial intelligence model that takes metadata and a question-and-answer with the user as input (S606). The second artificial intelligence model may include a natural language generation model.

[0113] “Completeness of user information” refers to an indicator of how faithfully user-related information required for writing a career description has been collected, and can be evaluated based on factors such as the specificity of responses, information entropy-based diversity indicators, results of applying personalized weights, and job matching by question type.

[0114] A user-customized career description generation device according to one embodiment of the present disclosure can generate a career description using a third artificial intelligence model based on metadata and question-and-answer with the user when the completeness of user information is above a preset standard (S608). The third artificial intelligence model may include a natural language generation model.

[0115] A user-customized career description generation device according to one embodiment of the present disclosure can provide the generated career description to the user through a user interface (S610).

[0116] A user-customized career description generation device can update the context vector using the user's additional response to an additional query.

[0117] If necessary, the user-customized career description generation device can evaluate the level of specificity regarding the user's response to additional queries. More specifically, the device may analyze the acquired user response to consider formal aspects, such as the length of the response sentence, the inclusion of quantitative numerical information like years of service, duration, amount, and number of personnel, the inclusion of proper nouns like company names, tool names, and project names, and the sentence structure of the response. It may also consider semantic aspects, such as the number of matches with the expected set of answer keywords and the extent to which Situation-Task-Action-Result (STAR) elements are reflected within the response. Furthermore, it may evaluate the specificity of the response by calculating the cosine similarity between the response vector and the ideal answer example vector through machine learning or language model-based analysis, or by directly scoring the specificity of the response on a numerical scale from 0 to 10 using a large-scale language model.

[0118] The specificity level calculated in this way can be classified, for example, as a short or ambiguous response in the range of 0 to 3, as a general description including some details in the range of 4 to 6, and as a specific description including quantitative figures, proper nouns, and detailed context in the range of 7 to 10. Accordingly, the device according to the present invention can quantitatively evaluate the specificity of a user response and, based on the result, proceed with the conversation by appropriately selecting a subsequent question, such as a drill-down question, an expansion question, or a validation question.

[0119] Accordingly, the device according to the present invention can quantitatively evaluate the specificity level of a user response and, if necessary, additionally perform a drill-down question or an expansion question.

[0120] A user-customized career description generation device can generate a detailed inquiry that further limits or requires details regarding the content mentioned in the response when the specificity level is below a threshold, and generate an expanded inquiry that explores surrounding circumstances or background information related to the response when the specificity level is above a threshold.

[0121] If necessary, the user-customized career description generation device may collect job descriptions. The device may generate additional queries regarding the user by providing at least one query, the user's response thereto, and the job description as input to a first artificial intelligence model. Additionally, the user-customized career description generation device may evaluate the degree of matching between the job description and the response to the additional query by providing metadata and the history of the query and response with the user as input to a second artificial intelligence model. If the degree of matching is below a preset standard, the user-customized career description generation device may generate additional supplementary questions corresponding to the unmatched requirements; if the degree of matching is above the preset standard, the device may generate a career description by prioritizing experiences that have a high degree of suitability to the job description. Here, the evaluation of the degree of matching may be performed based on at least one of the required competencies, required years of experience, technical keywords, and job suitability indicators listed in the job description.

[0123] As described above, exemplary embodiments have been disclosed in the drawings and specification. Although specific terms have been used to describe the embodiments in this specification, they are used only for the purpose of explaining the technical concept of this disclosure and are not intended to limit the meaning or the scope of this disclosure as defined in the claims. Therefore, those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of this disclosure should be determined by the technical concept of the appended claims.

Claims

Claim 1 A method for generating a user-customized career description performed by a program executed on a computing device, wherein the device comprises the steps of: collecting metadata regarding a user from a user; generating at least one query related to the user; generating a context vector by analyzing the at least one query and the user response thereto using a first artificial intelligence model, extracting key elements from the context vector, calculating a question generation probability distribution based thereon, constructing an answer branching tree for a plurality of candidate questions and calculating information entropy for the branches, calculating a priority score of the questions by applying a personalized weight based on the user's past response patterns, and selecting and generating the candidate question with the highest expected value of information acquisition based on the question generation probability distribution, information entropy, and personalized weight as an additional query regarding the user under a question diversity constraint; evaluating the specificity level of the user response to the additional query, generating a detailed inquiry that further limits the content mentioned in the response if the specificity level is below a threshold, and generating an extended inquiry that expands surrounding information related to the response if the specificity level is above a threshold; A method for generating a user-customized career description, comprising: a step of the device cumulatively updating the context vector by reflecting the user response to the additional query; a step of the device evaluating the completeness of user information using a second artificial intelligence model that takes the metadata and the query-response with the user as input, and repeating the additional query generation step and the context vector update step if the completeness is below a preset standard; and a step of the device generating a career description using a third artificial intelligence model that takes the metadata and the query-response with the user as input if the completeness of the user information is above a preset standard. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 A method for generating a user-customized career description according to claim 1, wherein the metadata includes information on at least one of the user's name, age, educational background, major, years of experience, job field, skills possessed, certifications, awards, and desired job. Claim 8 A method for generating a user-customized career description according to claim 1, further comprising: a step of collecting job descriptions; a step of generating additional queries regarding the user using a first artificial intelligence model that takes as input the at least one query, the user response to the at least one query, and the job description; a step of evaluating the degree of matching between the job description and the response to the additional query using a second artificial intelligence model that takes as input the metadata and the history of question-and-answer exchanges with the user; a step of generating supplementary questions for unmatched requirements if the degree of matching is below a preset standard; and a step of generating a career description by arranging experiences in order of high suitability with the job description if the degree of matching is above a preset standard. Claim 9 A method for generating a user-customized career description according to claim 8, wherein the matching evaluation is performed based on at least one of the required competencies of the job posting, required years of experience, technical keywords, and job suitability indicators. Claim 10 A user-customized career description generation device comprising: a processor including at least one core; a memory including program codes executable on said processor; metadata regarding a user, and a network unit for receiving a response from said user; and a user interface for displaying the generated career description.The processor includes: collecting metadata about a user from a user; generating at least one query related to the user; generating a context vector by analyzing the at least one query and the user response thereto using a first artificial intelligence model; extracting key elements from the context vector and calculating a question generation probability distribution based thereon; constructing an answer branching tree for a plurality of candidate questions and calculating information entropy for the branches; calculating a priority score for the questions by applying personalized weights based on the user's past response patterns; selecting and generating the candidate question with the highest expected value of information acquisition based on the question generation probability distribution, information entropy, and personalized weights as an additional query about the user under question diversity constraints; evaluating the specificity level of the user response to the additional query; generating a detailed inquiry that further limits the content mentioned in the response if the specificity level is below a threshold, and generating an extended inquiry that expands surrounding information related to the response if the specificity level is above a threshold; cumulatively updating the context vector by reflecting the user response to the additional query; and the metadata A device configured to evaluate the completeness of user information using a second artificial intelligence model that takes data and a query-response with a user as input, and if the completeness is below a preset standard, to repeatedly perform the additional query generation step and the context vector update step, and if the completeness of the user information is above the preset standard, to generate a career description using a third artificial intelligence model that takes metadata and a query-response with the user as input, and to provide the career description to the user through a user interface.

Images