Customized visa change and stay support system for foreign workers

Abstract

The present invention relates to a customized visa change and stay support system for foreign workers, and more specifically, to a comprehensive system comprising functions for AI-based job seeker-company matching, support for stay and visa change, review of employment contracts, wage payment, and administrative processing related to labor laws.

Description

Technology Field

[0001] The present invention relates to a customized visa change and stay support system for foreign workers, and more specifically, to a comprehensive system comprising functions for AI-based job seeker-company matching, support for stay and visa change, review of employment contracts, wage payment, and administrative processing related to labor laws. Background Technology

[0002] The current foreign worker employment system causes job search and visa-related information to be fragmented, making it difficult for foreign workers to efficiently find employment opportunities and maintain their residency status.

[0003] Under existing technologies, visa application and extension services, job information provision systems, and wage payment and employment contract management systems exist separately; however, technology that integrates these functions to provide AI-based optimized matching is lacking.

[0004] For example, Korean Registered Patent Publication No. 10-2421083 describes a system that provides an online agency service for foreign nationals' visa issuance and extension applications; however, this is a simple agency service and does not include a function that provides optimal matching by analyzing the worker's career history and the company's requirements using AI.

[0005] Furthermore, most existing job search platforms target Koreans, and there is no customized matching system that takes into account the legal residency requirements and visa change requirements of foreign workers.

[0006] Accordingly, there is a growing need for an integrated system that allows foreign workers to find optimal employment opportunities by comprehensively considering their residency status, work experience, and corporate requirements, while simultaneously receiving support for residency and labor legal services. The problem to be solved

[0007] The objective of the present invention is to provide an AI-based one-stop service system for supporting the employment and stay of foreign workers. Through this, the following problems are to be solved.

[0008] Unlike existing Korean-centric job search platforms, it recommends suitable companies by comprehensively analyzing the foreign worker's residency status (E-9, H-2, F-4, etc.) and corporate requirements.

[0009] It calculates the AI-based worker-company matching index (M) to recommend the most suitable workplace.

[0010] AI resolves residency issues for foreign workers by analyzing visa change requirements and automatically recommending necessary documents and procedures.

[0011] Efficient support is possible by linking with government administrative procedures (visa applications, tax filings, etc.).

[0012] It calculates an AI-based employment contract evaluation index (G) to determine the suitability of the contract and provides guidance on improvements to workers and companies.

[0013] It automatically detects delayed or unpaid wages and provides guidance on measures to prevent labor disputes.

[0014] Foreign workers face difficulties with loans and credit transactions due to a lack of financial history.

[0015] This system evaluates a worker's creditworthiness based on salary records, work history, etc., and supports the linkage of loan products.

[0016] We prevent unnecessary legal disputes by providing consulting services regarding tax filing, tax reductions, industrial accident insurance enrollment, and labor laws for foreign workers.

[0017] The problems of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0019] A customized visa change and stay support system for foreign workers according to an embodiment of the present invention for solving the above problem is a system that supports matching between foreign workers and companies, wherein

[0020] A worker terminal unit (300) that receives worker data including worker history information and mutual satisfaction;

[0021] A corporate representative terminal unit (400) receiving corporate data including corporate requirement specification information and mutual satisfaction; and

[0022] It is connected to the above worker terminal unit and the above company representative terminal unit via a network,

[0023] A data collection module (110) that receives the above worker data and the above company data and stores them in a database;

[0024] A calculation module (120) that calculates a final matching index by executing a first control logic, a second control logic, and a third control logic based on data stored in the above database; and

[0025] A result interpretation / notification module (130) that determines whether the match is suitable by comparing the calculated final matching index with a preset threshold value, and then transmits the determination result to the worker terminal unit and the company representative terminal unit;

[0026] A central server unit (100) including;

[0027] Includes,

[0028] The first control logic is characterized by calculating the final matching index by multiplying the objective fit index calculated by the second control logic and the subjective matching stability index calculated by the third control logic.

[0029] The above second control logic is,

[0030] A first operation of taking the natural logarithm of the value obtained by adding 1 to the square of the corporate evaluation score in which the above company evaluated the worker;

[0031] A second operation of taking the square root of the value obtained by adding 1 to the square of the difference between the resume suitability score of the above worker and the specification level required by the above company;

[0032] A third operation of adding 1 to the negative exponential function value of the value obtained by multiplying the physical distance between the said worker and the said company by a distance penalty mitigation factor calculated based on the frequency of mutual communication between the said worker and the said company;

[0033] Perform,

[0034] The objective fit index is calculated by dividing the sum of the result of the first operation and the result of the second operation by the result of the third operation.

[0035] The above third control logic is,

[0036] Calculate the difference between the employee satisfaction assigned to the company by the above-mentioned employee and the employee satisfaction assigned to the company by the above-mentioned company to the employee, and

[0037] It is characterized by calculating the subjective matching stability index by multiplying the difference value calculated above by a satisfaction deviation sensitivity coefficient and then performing a cosine function operation on the result value.

[0038] The above data collection module (110) is,

[0039] Receive worker data including the worker's corporate satisfaction from the worker terminal unit (300), and

[0040] Receives corporate data including the corporate evaluation score and the employee satisfaction of the company from the above corporate representative terminal unit (400), and

[0041] The address information of the above worker terminal unit and the above company representative terminal unit is transmitted to a map API to receive the above physical distance, and

[0042] The method is characterized by calculating the distance penalty mitigation coefficient by recording the number of message exchanges and interview requests between the worker terminal unit and the company representative terminal unit.

[0043] The above result interpretation / notification module (130) is,

[0044] If the above final matching index is greater than or equal to the first threshold, it is determined as 'strongly recommended matching' and a notification proposing a non-face-to-face interview schedule is sent to both terminal units, and

[0045] If the above final matching index is less than the first threshold and greater than or equal to the second threshold, it is determined as a 'matching consideration target' and additionally provides mutual profile information to both terminal units,

[0046] Characterized by determining 'unsuitable match' and lowering the mutual exposure ranking in the search results of both parties if the above final matching index is below the second threshold,

[0047] A manager terminal unit (200) for coordinating the operation policy of the system; and

[0048] A management UI module (140) that provides a graphical user interface to the above-mentioned administrator terminal unit and changes the scaling coefficient of the first control logic, the satisfaction deviation sensitivity coefficient of the third control logic, or the threshold value of the result interpretation / notification module (130) according to input from the above-mentioned administrator terminal unit;

[0049] It can be characterized by including more. Effects of the invention

[0051] The present invention is an AI-based integrated system that efficiently supports the employment and stay of foreign workers, and provides the following effects.

[0052] It provides optimal job matching by quantitatively evaluating workers' careers, qualifications, preferred conditions, and companies' requirements.

[0053] The compatibility between a workplace and an employee can be objectively assessed by utilizing the job seeker-company matching index (M).

[0054] Based on the user profile (nationality, residency status, work history), AI recommends the optimal visa change path.

[0055] By automating the visa application process, users can proceed with administrative procedures with just simple input.

[0056] It analyzes the suitability of employment contracts and automatically detects whether wage payments are delayed or overdue, providing a warning system.

[0057] Legal disputes can be prevented in advance by utilizing the employment contract evaluation index (G).

[0058] By utilizing a credit rating model for foreigners not provided by the existing banking system, it increases the availability of financial services (such as loans).

[0059] Expand loan access for foreign workers by partnering with P2P financial companies and banks.

[0060] Reduces the legal burden on foreign workers by automating administrative processing related to labor laws, taxation, and industrial accident insurance.

[0061] Reduces legal risks for companies and workers and creates a smooth working environment.

[0062] Reduces the problems of illegal employment and illegal residency by supporting the smooth employment and stay of foreign workers.

[0063] From a company's perspective, it is easy to find suitable foreign workers, which can resolve labor supply and demand issues.

[0064] New business models can be created through linkages with the financial sector, legal services, and HR markets.

[0065] The effects according to the present invention are not limited to those exemplified above, and a wider variety of effects are included within the present invention. Brief explanation of the drawing

[0067] FIG. 1 illustrates the overall relationship diagram of the present invention. Figure 2 schematically shows the relationship diagram of the server of the present invention. Specific details for implementing the invention

[0068] Hereinafter, various embodiments are described in more detail with reference to the attached drawings. The embodiments described in this specification may be modified in various ways. Specific embodiments may be depicted in the drawings and described in detail in the detailed description. However, specific embodiments disclosed in the attached drawings are intended only to facilitate understanding of various embodiments. Accordingly, the technical concept is not limited by specific embodiments disclosed in the attached drawings, and it should be understood that it includes all equivalents or substitutions that fall within the spirit and scope of the invention.

[0069] Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but these components are not limited by the aforementioned terms. The aforementioned terms are used solely for the purpose of distinguishing one component from another.

[0070] Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, if the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0071] The predefined rules of operation or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that a basic artificial intelligence model is trained using a number of training data by a learning algorithm, thereby creating predefined rules of operation or artificial intelligence models configured to perform desired characteristics (or objectives). Such learning may be performed on the device itself where the artificial intelligence according to the present disclosure is executed, or it may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

[0072] An artificial intelligence model can be composed of multiple neural network layers. Each of the multiple neural network layers has multiple nodes and weight values, and performs neural network operations through calculations between the results of previous layers and the multiple weights. The multiple weights possessed by the multiple neural network layers can be optimized based on the learning results of the artificial intelligence model. For example, multiple weights can be updated so that the loss value or cost value obtained by the artificial intelligence model during the learning process is reduced or minimized. Additionally, to minimize the loss value or cost value, multiple weights can be updated in a direction that minimizes the gradient associated with the loss value or cost value. Artificial neural networks may include deep neural networks (DNNs), such as Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs), Restricted Boltzmann Machines (RBMs), Deep Belief Networks (DBNs), Bidirectional Recurrent Deep Neural Networks (BRDNNs), or Deep Q-Networks, but are not limited to the examples mentioned above.

[0073] A network is a network that serves as a transmission path for web pages; it may be a closed network such as a LAN (Local Area Network) or WAN (Wide Area Network), but it is desirable for it to be an open network such as the Internet. The Internet refers to a global open computer network structure that provides the TCP / IP protocol and various services existing at its upper layers, namely HTTP (HyperText 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), and NIS (Network Information Service).

[0074] Terminals can be implemented in various forms. For example, the terminals described in this specification may include mobile terminals such as smartphones, tablet PCs, PDAs, portable multimedia players, and MP3 players, as well as fixed terminals such as smart TVs and desktop computers.

[0075] In this specification, terms such as “comprising” or “having” are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, 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. When a component is described as being “connected” or “connected” to another component, it should be understood that it may be directly connected to or connected to that other component, or that there may be other components in between. On the other hand, when a component is described as being “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

[0076] Meanwhile, a "module" or "part" for a component as used in this specification performs at least one function or operation. Furthermore, a "module" or "part" may perform a function or operation by hardware, software, or a combination of hardware and software. Additionally, a plurality of "modules" or a plurality of "parts," excluding a "module" or "part" that must be performed on specific hardware or on at least one processor, may be integrated into at least one module. A singular expression includes a plural expression unless the context clearly indicates otherwise.

[0077] In addition, power, power transmission, and control therefor for the following assembly configurations and embodiments, including "by control," follow conventional technology including terminals, applications, hardware control modules, etc., so they are omitted to avoid redundancy.

[0078] In addition, the operation embodiments and configurations described in a general manner without being explained in detail below follow the prior art and are omitted in order to focus on describing the purpose of the present invention and the resulting effects.

[0079] Furthermore, in describing the present invention, if it is determined that a detailed description of related known functions or configurations may unnecessarily obscure the essence of the invention, such detailed description is abbreviated or omitted.

[0080] In one embodiment, a customized visa change and stay support system for foreign workers may include a central server (100), a data collection module (110), a calculation module (120), a result interpretation / notification module (130), a management UI module (140), an administrator terminal (200), a worker terminal (300), and a company representative terminal (400).

[0081] The hardware components consist of a central server (100), an administrator terminal (200), a worker terminal (300), and a company representative terminal (400).

[0082] The definitions, functions, and data transmission and reception relationships of each hardware unit and software module constituting the system of the present invention are as follows.

[0083] The central server unit (100) is physical server hardware that performs the central role of the present invention and is configured to include a high-performance CPU, a large-capacity RAM, an SSD-based database storage device, and a gigabit network interface. In addition, it drives the software modules (110, 120, 130, 140) of the present invention installed internally and directly performs all operations according to the first to twelfth control logic. In addition, it stores and manages all data received from each terminal unit in a database and transmits the results of operations back to each terminal unit.

[0084] The administrator terminal unit (200) is a physical device used by a system administrator and can be implemented in the form of a desktop PC with a web browser installed or a tablet PC with a dedicated management application installed. Additionally, it connects to the management UI module (140) of the central server unit (100) through a security authentication procedure (ID, PW, OTP, etc.). The administrator monitors the overall operational status of the system through a graphical user interface (GUI) and inputs commands to adjust various coefficients and threshold values ​​used in the control logic.

[0085] The worker terminal unit (300) and the company representative terminal unit (400) are personal physical devices used by a foreign worker and a company recruiter, respectively, and are generally implemented as a smartphone, tablet PC, or laptop computer that accesses the system via a web browser, on which the dedicated application of the present invention is installed. Additionally, the user inputs data required for the control logic, such as personal history information, satisfaction scores, and attendance records, according to the input form of the application or web page installed on each terminal unit. The input data is transmitted to the central server unit (100) via an encrypted communication protocol (e.g., HTTPS), and result values, such as the matching index and contract grade received from the central server unit (100), are displayed on the screen.

[0086] The data collection module (110) is a software component responsible for receiving all raw data required for control logic operations from each terminal unit, processing it according to a defined format, and storing it in a database. For example, it receives satisfaction data such as { "user_id": "workerA", "satisfaction_score": 8} in JSON (JavaScript Object Notation) format from a worker terminal unit (300) through an API (Application Programming Interface) endpoint. In addition, it parses the received data to verify the user ID and inserts or updates the satisfaction score 8 into the 'satisfaction' table of the database in the row of the corresponding user.

[0087] The calculation module (120) is a software computation engine that actually executes the control logics (first to twelfth) of the present invention. Additionally, it retrieves values ​​stored in a database by the data collection module (110). For example, to perform the first control logic, all necessary variable values, such as corporate evaluation scores and resume suitability scores, are loaded into memory, and then the final matching index is calculated according to the order of operations defined in the specification.

[0088] The result interpretation / notification module (130) is a software component that interprets a numeric result (e.g., final matching index 14.62) calculated by the calculation module (120), converts it into meaningful information (e.g., 'matching unsuitable') for the user, and notifies the user of this through an appropriate channel. For example, it internally receives a calculation result such as {"match_id": "M123", "final_score": 14.62} from the calculation module (120). In addition, it generates a judgment result of 'matching unsuitable' by comparing it with a pre-set threshold (40). Afterward, it constructs a push notification payload containing the judgment result and sends it to the worker terminal unit (300) and the corporate representative terminal unit (400).

[0089] The management UI module (140) is a software interface that provides a visual dashboard screen for the administrator to the administrator terminal unit (200) and receives input commands from the administrator to change the settings of other modules or databases of the central server unit (100). Additionally, when the administrator enters 1.5 into the scaling factor input field on the web screen of the administrator terminal unit (200) and clicks the 'Save' button, an HTTP POST request such as {"param": "scaling_factor", "value": 1.5} is received. Additionally, in accordance with the received request, an SQL query is executed to update the scaling factor value to 1.5 in the 'system settings' table within the database.

[0090] The matching system of the present invention calculates a final matching index between a worker and a company through a first control logic, a second control logic, and a third control logic.

[0091] The final matching index is a final indicator that quantifies the overall suitability between workers and companies into a value between 0 and 100, and is calculated by reflecting a combination of objective conditions and subjective satisfaction.

[0092] The objective fit index is an indicator calculated by combining the objective capabilities of workers, the requirements of the company, and the physical constraints between the two.

[0093] The subjective matching stability index is an indicator of how well subjective satisfaction matches between workers and companies. It has a value between 0 and 1, and the closer it is to 1, the more perfectly mutual satisfaction matches.

[0094] The scaling factor is a coefficient that adjusts the overall magnitude of the final matching index, and its purpose is to correct the sensitivity of the matching index according to industry group or country-specific characteristics.

[0095] The company evaluation score is an objective competency score evaluated by the company through the employee's documents, interviews, technical tests, etc., and has a value between 0 and 100 points.

[0096] The resume suitability score is a score automatically calculated by the system based on the resume information submitted by the worker according to a predefined standard table, and has a value between 0 and 50 points.

[0097] The level of specifications required by the company is a score quantified by the system based on the qualifications, experience, and other requirements specified by the company in the job posting, and has a value between 0 and 50 points.

[0098] The distance penalty mitigation factor is a coefficient calculated based on the frequency of mutual communication between workers and companies, has a value between 0 and 1, and determines the degree to which the reduction effect of the matching index due to physical distance is mitigated.

[0099] Physical distance is the actual distance between the worker's address and the company's location converted into kilometers (km), with an effective range between 0km and 1000km.

[0100] Employee satisfaction with the company is a subjective satisfaction score in which employees evaluate the company's working environment, welfare, vision, etc., and has a value between 1 and 10 points.

[0101] Employee satisfaction is a subjective satisfaction score in which the company evaluates employees' attitudes, communication skills, etc., and has a value between 1 and 10 points.

[0102] The satisfaction deviation sensitivity coefficient is a coefficient that moderates the effect of mutual satisfaction deviation on the subjective matching stability index.

[0103] The prior art used a simple model that independently evaluated each element and linearly summed them when matching workers and companies.

[0104] This had a clear limitation in that it failed to reflect the reality of complex interactions occurring in the actual hiring process, such as the fact that even if objective conditions are perfect, a match can fail if subjective satisfaction is low.

[0105] To overcome these limitations, the present invention defines two axes, 'objective fitness' and 'subjective stability,' and uniquely proposes a 'multiplicative interaction model' in which these two axes influence each other.

[0106] The first control logic calculates the final matching index through the following operation process.

[0107] The final matching index is calculated by multiplying the objective goodness-of-fit index and the subjective matching stability index, and then additionally multiplying by a scaling factor.

[0108] The second control logic calculates the objective fitness index through the following operation process.

[0109] The objective fit index is calculated by dividing the sum of the natural logarithm of the square of the corporate evaluation score plus 1, the square root of the difference between the resume fit score and the level of specifications required by the company plus 1, and the negative exponential function of the product of physical distance and the distance penalty mitigation factor plus 1.

[0110] Each operation constituting the second control logic described above is designed with technical necessity to precisely model the complex and non-linear laws of reality observed in the adoption market.

[0111] First, the first operation on the score in which a company evaluates a worker is a key technical means of modeling the 'law of diminishing marginal value of highly skilled talent.' Linear score models of prior art have a limitation in that they treat the difference in value between a score increase in the low range and a score increase in the high range as equal. However, in reality, whether a person possesses basic competencies has a greater impact on matching than the minute score differences among highly skilled talents above a certain level. The operation in the form of a natural logarithm adopted in this invention implements the characteristic that the rate of increase of the result value gradually slows down as the input evaluation score increases. This has a technical stabilization effect that prevents a small number of ultra-high-scoring talents from distorting the score distribution of the entire matching system and provides fair opportunities to a large number of candidates who possess competencies above a certain level. Furthermore, the structure of squaring the input evaluation score, adding a certain positive number, and then taking the logarithm is the result of engineering considerations for a robust system design that emphasizes only the positive influence of the score, while fundamentally preventing calculation errors that may occur when the input score is zero or negative.

[0112] Second, the second operation regarding the difference between the worker's resume suitability score and the company's required specification level models the concept of a 'discrepancy tolerance threshold.' In the hiring process, perfect alignment between a worker's capabilities and the company's required specifications is rare, and it is common to allow for a certain degree of discrepancy. Using a simple subtraction method, as in the prior art, presents a problem where even minor specification discrepancies cause the matching score to drop significantly, potentially leading to the loss of excellent matching opportunities. The square root-type operation of the present invention mitigates penalties in ranges where the difference between the two scores is small, allowing for a lenient evaluation of minor discrepancies, while gradually stabilizing the impact in ranges where the difference is large to prevent excessive penalties. This mathematically implements a realistic evaluation method used by hiring managers, which responds flexibly to minor specification differences while imposing clear penalties on serious discrepancies.

[0113] Third, the third operation regarding the physical distance between the worker and the company models not a simple distance penalty, but the 'phenomenon of overcoming psychological distance.' While physical distance is a clear constraint, active communication between the two parties (exchange of messages, requests for interviews, etc.) can be interpreted as a positive signal to overcome this constraint. The present invention reflects this dynamic interaction as a key variable called the 'distance penalty mitigation coefficient.' The operation in the form of a logistic function including the aforementioned negative exponential function creates a non-linear effect in which the probability of matching decreases sharply with distance when there is no communication at all between the two parties, but the penalty is significantly mitigated even at the same distance as communication becomes more active. This is a unique technical solution of the present invention that transforms the constraint of 'static distance' into the 'result of dynamic interaction,' technically realizing aspects of human interaction that prior art fails to capture.

[0114] In conclusion, the second control logic is a technical means that calculates objective suitability by synthesizing multidimensional information through the organic combination of first, second, and third operations that model different laws of reality.

[0115] The third control logic calculates the subjective matching stability index through the following operation process.

[0116] The subjective matching stability index is calculated by performing a cosine function operation on the value obtained by multiplying the difference between employee satisfaction with the company and employee satisfaction with the company by the satisfaction deviation sensitivity coefficient.

[0117] The application of a cosine function-type operation to the aforementioned third control logic is not merely a simple mathematical device, but a unique technical solution of the present invention for quantitatively modeling the phenomenon of 'relational resonance' or 'harmony' between two entities, which the linear difference value model of the prior art can never capture.

[0118] Prior art uses methods that simply subtract the difference in satisfaction between the two parties or take the absolute value. This linear approach has a clear limitation in that it provides only information on "how much the difference is" and fails to reflect the "direction of the difference" or the "qualitative state of the relationship." For example, a small difference when both the worker and the company are highly satisfied and an identical difference when both are dissatisfied imply that the quality of the relationship is fundamentally different.

[0119] To overcome these limitations, the present invention regards the difference in satisfaction as a concept similar to the 'phase difference' of a wave and applies an operation in the form of a cosine function. According to the above model, the following effects occur.

[0120] First, it is possible to model a state of perfect resonance. When the satisfaction levels of employees and the company align perfectly with no difference, the subjective matching stability index reaches its maximum value. This signifies that the relationship is most stable because the expectations and evaluations of both parties are in perfect harmony.

[0121] Second, it is possible to model dissonance and cancellation effects. As the difference in satisfaction gradually widens, the stability index value gradually decreases, much like two waves with different phases canceling each other out. In particular, when the difference in satisfaction reaches a certain threshold, the stability index converges to zero, which mathematically expresses that the power of the matching is completely extinguished due to subjective instability, no matter how good other objective conditions are.

[0122] Third, it is possible to model negative interactions in the relationship. When the difference in satisfaction becomes extreme, leading to the worst-case scenario, the stability index takes on a negative value. This models an 'adversarial relationship' where both parties rate each other the worst possible, indicating that a force actively hindering matching is at work. This cannot be implemented using a simple difference value model.

[0123] Furthermore, the 'satisfaction deviation sensitivity coefficient' included in the third control logic is a key technical parameter that regulates the sensitivity of this resonance phenomenon. This coefficient plays a role in adjusting the importance of 'subjective harmony' on overall matching stability depending on the characteristics of the industry or job function. For example, in jobs where teamwork is critical, this coefficient value can be increased to enable a sensitive response to even small differences in satisfaction, while in independent work environments, the value can be lowered to provide flexibility. This coefficient is designed to allow system operators to actively adjust it via the administrator terminal unit, enabling system operation optimized for various recruitment environments.

[0124] In conclusion, the third control logic is a highly sophisticated psychometric model that goes beyond the simple comparison of satisfaction scores to quantify the qualitative state of interaction between two subjects by likening it to a physical phenomenon called 'resonance,' thereby predicting the stability of the relationship and the potential for conflict. This is a technical feature that differs in dimension from the one-dimensional score calculation method of prior art.

[0125] The core of the present invention lies in combining the calculated objective fit index and subjective matching stability index through multiplication, rather than simply adding them together.

[0126] This multiplicative combination method is not merely a modification of mathematical operators, but is the main technical concept of the present invention designed to overcome the fundamental technical limitations of the linear summation model of the prior art and to more accurately model the complex matching processes of reality.

[0127] Specifically, the simple summation evaluation model used in the prior art is based on the assumption that each evaluation element is independent, and thus contains a fatal limitation in that it fails to properly reflect 'critical disqualifications.' For example, even if a worker's objective competence (objective fit index) is perfect enough to be close to a perfect score of 100, a match in reality can never succeed if their subjective satisfaction with the corporate culture or job (subjective matching stability index) is close to 0. However, the summation model of the prior art still produces a high score above the intermediate level even in such cases, causing a serious problem of information distortion by presenting a match that is clearly destined to fail to both the user and the company as promising.

[0128] To solve these problems, the present invention adopts multiplication operations to technically implement the 'law of the minimum of matching' or the 'principle of chain combination,' in which each evaluation indicator influences the effect of others. This mathematically models the real-world principle that the success of matching is determined by the weakest factor. According to the multiplicative combination of the present invention, if either the objective condition or the subjective stability index converges to zero, the final matching index is inevitably calculated to be close to zero, regardless of how high the other index is.

[0129] Accordingly, the first control logic of the present invention goes beyond simply listing or summing individual scores as in the prior art and comprehensively evaluates the 'solidity of relationships' by reflecting the organic interdependence between each element. This creates a remarkable and heterogeneous effect that could never have been predicted by a simple combination of conventional technologies, which dramatically increases the success rate of matching and prevents the user from wasting time by filtering out 'critical disqualifying factors' early and recommending only stable matches that satisfy minimum standards in all aspects.

[0130] In addition, the 'scaling coefficient' included in the first control logic is a technical means for adjusting the overall sensitivity of this multiplication model. It serves to optimally correct the final distribution of the matching index according to the characteristics of the industry group (e.g., production jobs with frequent turnover vs. research jobs where long-term employment is important) or the labor market of each country, and ensures the flexibility and adaptability of the system by allowing the system operator to actively adjust it through the manager terminal unit (200).

[0131] This multiplicative combination results in a sharp decline in the final matching index if stability is reduced due to mutual subjective dissatisfaction, even if objective conditions are perfect.

[0132] This mathematically embodies the complex real-world decision-making process where "successful matching is impossible if either objective conditions or subjective stability are severely lacking," providing unpredictable synergistic effects that can never be achieved through a simple combination of prior art technologies.

[0133] Each variable of the present invention is collected in the following specific and reproducible manner, excluding the user's arbitrary judgment.

[0134] When a company representative inputs 80 points for the technical interview result and 90 points for the personality interview result for worker A through the interface of the company representative terminal (400), the system calculates the final company evaluation score as 80*0.6 + 90*0.4 = 84 by applying a preset weight (technical 60%, personality 40%), and transmits this value to the central server (100) for storage.

[0135] When worker B inputs information such as 'possession of a master's degree in a related field' (15 points), '4 years of experience in a related job' (10 points), and 'possession of one required certification' (10 points) through the worker terminal (300), the central server (100) automatically calculates the resume suitability score as 15+10+10 = 35 by summing the scores of each item according to a standard table such as [Table 1] of the specification.

[0136] When Company C sets the hiring conditions as ‘master’s degree or higher’ (15 points), ‘3 years or more of experience’ (10 points), and ‘2 or more required certifications’ (15 points) on the company representative terminal (400), the central server (100) automatically calculates the company’s required specification level by summing the corresponding standard scores to 15+10+15 = 40.

[0137] The physical distance is the address (Gangnam-gu, Seoul) received from worker D's terminal (300) and the address (Bundang-gu, Seongnam-si, Gyeonggi-do) received from company E's terminal (400) transmitted to the map API, and the shortest distance on the returned path, 15.3, is stored as the physical distance.

[0138] The process of calculating the above distance penalty mitigation factor does not simply involve substituting values ​​into a fixed formula, but rather follows the following step-by-step logical procedure.

[0139] Step 1: In the 'Communication Proactiveness Index' calculation step, the central server (100) analyzes the interaction records between employee D and company E. In this process, general message exchanges (12 times) and the company's official video interview request (1 time) exchanged between the two parties are distinguished and assigned different weights to each. In particular, the act of requesting an official interview is intentionally given a much higher score than the act of simple message exchange, in order to technically model the real-world interaction in which an interview request indicates a stronger and more serious intention to hire in order to overcome physical distance. The scores for each interaction calculated in this way are summed to calculate the final 'Communication Proactiveness Index'.

[0140] Step 2: In the 'perceived distance disadvantage' calculation step, it is determined that the higher the 'communication proactiveness index' mentioned above, the less negative impact the constraint of physical distance has on the matching. The central server (100) calculates a 'distance mitigation value' to be subtracted from the actual physical distance based on the index, and through this, calculates the final 'perceived distance disadvantage' that both parties psychologically feel.

[0141] Step 3: Based on the 'perceived distance disadvantage' finally calculated in the 'Distance Penalty Mitigation Factor' determination step, the 'Distance Penalty Mitigation Factor' to be finally reflected in the matching index calculation is determined.

[0142] Mutual satisfaction (worker / company) is stored in the central server (100) when, after the interview ends, worker F enters the worker’s satisfaction with company G as 9 points on the worker terminal (300) saying ‘I like the work environment and growth potential,’ and the person in charge of company G enters the company’s worker satisfaction as 8 points on the company representative terminal (400) saying ‘I am impressed by F’s proactive attitude.’

[0143] The calculated final matching index is used to determine the goodness of fit of the match through comparison with a pre-set threshold.

[0144] For example, three threshold levels can be set through the administrator terminal (200) as follows.

[0145] Active matching recommendation (final matching index ≥ 70) initiates an active matching process, such as the system automatically suggesting a non-face-to-face interview schedule to the relevant worker and company.

[0146] Matching considerations (40 ≤ final matching index < 70) recommend that both parties add each other to their 'interest list' and provide additional information (e.g., portfolio, corporate welfare information).

[0147] Incompatible matches (final matching index < 40) lower each other's exposure rankings in the search results of both parties or recommend other more suitable targets.

[0148] Explain the matching process between worker A and company B using an example.

[0149] The input values ​​are as follows:

[0150] Scaling Factor=1.5, Company Rating Score=85, Resume Fit Score=40, Company Required Spec Level=35, Distance Penalty Mitigation Factor=0.6, Physical Distance=50, Worker Company Satisfaction=8, Company Worker Satisfaction=6, Satisfaction Deviation Sensitivity Factor=0.4

[0151] Step 1 is the calculation of the objective fitness index (second control logic). For the numerator calculation, the natural logarithm of the square of 85 plus 1 is approximately 8.88, and the square root of the square of the difference between 40 and 35 plus 1 is approximately 5.10. The sum of the two values ​​is approximately 13.98.

[0152] To calculate the denominator, adding 1 to the negative exponential function value of the product of 0.6 and 50 gives approximately 1.0.

[0153] To calculate the objective goodness-of-fit index, 13.98 is divided by 1.0, and 13.98 is obtained.

[0154] Step 2 is the calculation of the subjective matching stability index (third control logic), and the result of the cosine function operation on the difference between 8 and 6 multiplied by 0.4 yields approximately 0.697.

[0155] Step 3 is the calculation of the final matching index (first control logic), in which the scaling factor 1.5, the objective goodness of fit index 13.98, and the subjective matching stability index 0.697 are all multiplied to calculate the final matching index to be approximately 14.62.

[0156] In step 4, the final matching index of 14.62 calculated as the result judgment is less than the threshold of 40, so it is judged as 'unsuitable match'.

[0157] This means that although objective conditions were relatively favorable, stability was lowered due to the large difference in subjective satisfaction between the two sides, resulting in a significant decrease in the final matching index.

[0158] The control logics of the present invention are designed to be robust to changes in input values ​​and to ensure a stable output.

[0159] The natural logarithm term of the second control logic prevents calculation errors through a structure that adds 1 to the input value, and ensures system stability by ensuring that the output increases gradually even as the input value increases.

[0160] The denominator term of the second control logic also fundamentally prevents division by zero through a structure that adds 1.

[0161] The cosine function of the third control logic always limits its output to between -1 and 1, thereby preventing differences in subjective satisfaction from abnormally affecting the final score.

[0162] [Table 1]

[0163]

[0164] The input variables used in the present invention are calculated based on objective and measurable data.

[0165] In particular, the resume suitability score is automatically calculated by the central server (100) according to a specific scoring criteria table such as the example [Table 1].

[0166] The distance penalty mitigation factor is also converted into a predefined formula based on objective data, namely the number of communications measured within the system.

[0167] The operation of taking the natural logarithm of the square of the corporate evaluation score plus 1, rather than a simple score, is intended to model the law of reality that 'the value of highly skilled talent increases non-linearly.' The 5-point difference between 90 and 95 points carries a much greater value than the 5-point difference between 50 and 55 points, and the above operation is a technical means to precisely reflect this difference.

[0168] The distance penalty mitigation factor plays a key role in transforming the static constraint of physical distance into the outcome of dynamic interaction. This technically realizes an aspect of human interaction—that distance can be overcome through psychological effort—that prior art completely failed to consider.

[0169] Modeling the deviation in mutual satisfaction using a cosine function is intended to mathematically express the phenomenon of 'harmony' or 'resonance' between two entities. The effect in which the power of matching is fully transmitted when satisfaction aligns (when the phases are the same) and canceled out when it is misaligned (when the phases are different) is a technical effect unique to this invention that cannot be realized through simple calculation of difference values.

[0170] Terms such as "optimization" or "minimization" used in this specification have clear criteria as follows.

[0171] In the present invention, "can be optimized" means a process in which a central server (100) uses past matching data (input variables and actual hiring success status) of 10,000 or more as training data to adjust a scaling coefficient and a satisfaction deviation sensitivity coefficient. At this time, the goal of optimization is to make the Pearson correlation coefficient between the calculated final matching index and the actual hiring success rate greater than or equal to a pre-set target value of 0.80.

[0172] "Minimizing" the loss value means calculating the root mean square error between the prediction result of the artificial intelligence model and the actual result, and using a gradient descent algorithm to converge the root mean square error value to a target threshold of 0.05 or less.

[0173] In addition, the system of the present invention calculates a comprehensive rating index through the fourth control logic, the fifth control logic, and the sixth control logic to quantitatively evaluate the level of fulfillment of the employment contract.

[0174] This aims to resolve the problem where the final rating is distorted because the prior art uses a linear model that evaluates each item of a contract individually and simply sums them up, causing high scores for other items even if a single fatal defect (e.g., unpaid wages) exists.

[0175] To overcome this, the present invention proposes a 'multiplicative reliability model' in which the substantive soundness of the contract and the sincerity of actual performance mutually influence each other, thereby comprehensively determining the reliability of contract performance.

[0176] Each indicator used in calculating the overall rating index and its purpose are as follows.

[0177] The overall rating indicator is a final reliability indicator calculated as a value between 0 and 100 by combining the substantive soundness of the contract and the sincerity of actual performance.

[0178] The contract soundness index is an index indicating how faithfully the employment contract itself meets legal requirements, and it has a value between 0 and 1.

[0179] The performance integrity index is an indicator of how faithfully a company has actually performed the contents specified in a contract. It has a value between 0 and 1, and the closer it is to 1, the more perfect the performance integrity is.

[0180] The grade scaling coefficient is a coefficient that adjusts the range of the final value of the comprehensive grade index, and is based on setting it to 100 to convert it into a percentage score.

[0181] The contract compliance importance coefficient is a coefficient that adjusts the weight of the contract soundness index, and has a value between 0 and 5, with higher values ​​set for industry groups with greater legal risk.

[0182] The qualitative indicator of the employment contract is a value calculated on a scale of 0 to 100 by the system automatically analyzing the degree of compliance with the essential clauses of the employment contract.

[0183] The wage payment time compliance importance coefficient is a coefficient that controls the sensitivity of the impact of wage delays on performance integrity, and has a value between 0 and 5.

[0184] The number of days of delayed wage payment is the number of days the actual wage is delayed compared to the payment date specified in the contract, and has an integer value greater than or equal to 0.

[0185] The wage ratio consistency coefficient is a coefficient that controls the impact of the difference between the ratio of actual paid wages and contractual wages on performance integrity. It has a value between 1 and 10, and the higher the value, the more sensitive it is to even slight discrepancies.

[0186] The actual pay / contractual wage ratio is the ratio of the monthly salary actually paid to the amount specified in the contract.

[0187] The working hours compliance indicator is a value calculated on a scale of 0 to 10 by the system evaluating the degree of compliance with statutory working hours per week.

[0188] Prior art remained a linear evaluation model that simply summed up individual scores for contract quality, wage payment, and compliance with working hours.

[0189] This model contains a logical contradiction in that even when the contract quality is 100 points and the wage payment score is 0 points, an intermediate score can still be produced.

[0190] Based on the 'Law of the Minimum,' which states that the reliability of a contract is determined by its weakest point, this invention developed a new evaluation model in which each evaluation element is connected through multiplication.

[0191] The fourth control logic calculates the overall rating index through the following calculation process.

[0192] The overall rating index is calculated by multiplying the contract soundness index and the performance integrity index, and then additionally multiplying by a rating scaling factor.

[0193] The 5th control logic calculates the contract soundness index through the following calculation process.

[0194] The contract soundness index is calculated by multiplying the contract compliance importance coefficient by the square root of the employment contract quality indicator and dividing the maximum value (5) of the contract compliance importance coefficient by the square root of the maximum value (100) of the employment contract quality indicator.

[0195] The 6th control logic calculates the fulfillment integrity index through the following operation process.

[0196] The compliance integrity index is calculated by multiplying the wage delay penalty index, the wage ratio penalty index, and the working hours compliance index.

[0197] At this time, the wage delay penalty index is calculated through a negative exponential operation on the product of the wage payment time compliance importance coefficient and the value obtained by dividing the wage payment delay days by 7, the wage ratio penalty index is calculated through a negative exponential operation on the value obtained by multiplying the square of the actual payment / contractual wage ratio minus 1 by the wage ratio consistency coefficient, and the working time compliance index is calculated by dividing the working time compliance indicator by the maximum value (10) of the indicator.

[0198] All variables used in the control logic of the present invention are automatically calculated from objective data rather than from the user's arbitrary judgment.

[0199] The qualitative indicator of the employment contract is calculated by the natural language processing module of the central server (100) analyzing the contract PDF file uploaded from the company representative terminal (400), confirming that it contains 4 of the 5 mandatory items specified in 'Article 17 of the Labor Standards Act', and automatically calculating the qualitative indicator of the employment contract as 80 points (4 / 5) * 100 = 80.

[0200] The number of days of delayed wage payment and the ratio of actual payment to contractual wage are calculated by the central server (100) linking with the company's salary transfer system, confirming that the salary was transferred on the 30th, which is 5 days later than the contractual payment date of the 25th of every month, and storing the number of days of delayed wage payment as 5, and confirming that the actual amount transferred is 2.97 million won compared to the contractual wage of 3 million won, and automatically calculating the ratio of actual payment to contractual wage as 297 / 300 = 0.99.

[0201] The indicator of compliance with working hours is automatically assigned by the central server (100) in conjunction with the attendance record app data of the worker terminal (300) to confirm that the average weekly working hours for one month have not exceeded the legal standard of 40 hours, and the indicator of compliance with working hours is 9 points (maximum 10 points).

[0202] The calculated comprehensive rating indicator is utilized according to clear criteria as follows.

[0203] The Best Contract (Overall rating index ≥ 80) certifies the company as a 'Trusted Company' and provides the benefit of simplifying some of the document screening procedures when hiring foreign workers in the future.

[0204] Good contracts (50 ≤ Comprehensive Rating Index < 80) are maintained as standard contracts.

[0205] For cautionary contracts (30 ≤ overall rating index < 50), the system automatically sends a notification to the company representative containing specific items requiring improvement, such as "reliability decline due to delayed wage payment."

[0206] Risk contract (overall rating index < 30) is automatically sent by the system to the administrator terminal (200) as a ‘warning of potential labor dispute’ and is provided as reference material that can link the company to the Ministry of Employment and Labor’s labor inspection system.

[0207] The process for calculating the contract performance reliability of Company C is as follows.

[0208] The input values ​​are as follows:

[0209] Grade Scaling Factor=100, Contract Compliance Importance Factor=3, Employment Contract Qualitative Indicator=81, Wage Payment Time Compliance Importance Factor=4, Days of Wage Payment Delay=10, Wage Ratio Consistency Factor=5, Actual Payment / Contractual Wage Ratio=0.9, Working Hours Compliance Indicator=7

[0210] Step 1 is the calculation of the contract soundness index (5th control logic), whereby the value (27) obtained by multiplying the square root (9) of 3 and 81 is divided by the value (50) obtained by multiplying the square root (10) of 5 and 100, and the contract soundness index is calculated as 0.54.

[0211] Step 2 is the calculation of the performance integrity index (6th control logic), and the result of the negative exponential function operation for the product of 4 and (10 / 7) for the wage delay penalty index is approximately 0.0033.

[0212] For the wage ratio penalty index, the result of the negative exponential operation of the square of (0.9-1) multiplied by 5 yields approximately 0.9512.

[0213] For the working hours compliance index, 7 is divided by 10 to obtain 0.7.

[0214] By multiplying the performance integrity index by 0.0033, 0.9512, and 0.7, the performance integrity index is calculated to be approximately 0.0022.

[0215] Step 3 is the calculation of the final comprehensive rating index (4th control logic), in which the rating scaling factor of 100, the contract soundness index of 0.54, and the performance integrity index of 0.0022 are all multiplied to calculate the comprehensive rating index to be approximately 0.1188.

[0216] In step 4, the overall rating index of 0.1188 calculated as the result judgment is less than the threshold of 30, so it is determined to be a 'risky contract'.

[0217] This clearly shows that although the contract itself was not bad (qualitative indicator of employment contract = 81), a 10-day delay in wages fatally lowered the integrity of performance and ultimately destroyed the credibility.

[0218] The use of an exponential function when evaluating wage delays is intended to model the 'speed of trust collapse.' It accurately reflects the non-linear reality where the decline in trust is gradual when the delay is only one or two days, but collapses exponentially once it exceeds a week. This is an effect that cannot be achieved using a linear penalty method.

[0219] When evaluating the wage ratio, using a negative exponential function of the square of the actual payment / contractual wage ratio minus 1 is intended to measure 'contractual consistency.' Penalties are imposed in both directions, considering not only cases where wages are paid less than stipulated in the contract but also cases where they are paid excessively without agreement as abnormal situations. This technology goes beyond the prior art of simple non-payment penalty methods and evaluates the predictability and transparency of contract performance.

[0220] Connecting the performance integrity index and the contract soundness index through multiplication embodies the principle of 'solidarity of trust.' Just as a contract becomes useless if it is not actually executed, no matter how perfect it may be, a serious defect in either party creates an effect that collapses the overall trust. This is the most fundamental difference distinguishing the present invention from the prior art, which treated each element independently.

[0221] In this specification, "optimization" of variables (contract compliance importance coefficient, wage payment time compliance importance coefficient, wage ratio consistency coefficient) means a process of adjusting each coefficient in a direction that makes the probability of the overall rating index value of a contract that has had a dispute predicted to be less than 30 points, which is the risk contract threshold, 95% or higher, by using 5,000 contract data cases of labor disputes accumulated over the past year as training data.

[0222] To overcome the limitation of the prior art, which merely calculates the absolute score of individual matching pairs, the system of the present invention includes additional control logic that evaluates the rationality of the matching by uniquely introducing the economic concept of 'opportunity cost'.

[0223] This fundamentally improves the quality of matching decisions by providing an answer to the question of whether the current choice is better than all other alternatives that must be forgone, going beyond simply asking whether the choice is good.

[0224] The opportunity cost-based matching system of the present invention calculates an opportunity cost-based final matching index through the 7th control logic, the 8th control logic, and the 9th control logic.

[0225] The opportunity cost-based matching index is a relative value indicator calculated by dividing the 'expected utility' of the current matching pair by the 'average opportunity cost' of the market, and the greater the value than 1, the more economically rational the current choice is.

[0226] The current choice expected utility is the total amount of net value expected to be created by a match between a specific worker currently under consideration and a specific company.

[0227] The market average opportunity cost is the sum of the average expected utility of all other firms that a particular worker could have chosen and the average expected utility of all other workers that a particular firm could have hired, and represents the average value of the alternatives forgone by the current choice.

[0228] Competency-required specification fit is an indicator that quantifies the degree of agreement between a worker's objective competencies (company evaluation score, resume fit score) and the company's required specification level as a value between 0 and 1.

[0229] Intersubjective synergy is a subjective value amplification coefficient calculated based on the mutual satisfaction of workers and companies, and has a value greater than 1 as mutual satisfaction increases.

[0230] Physical constraint costs are a quantified value of tangible and intangible costs resulting from physical distance and the lack of mutual communication.

[0231] Prior art had a fundamental limitation in that it merely calculated an isolated 'score' and could not compare how much value that score held within the entire market.

[0232] The present invention presents a new technical feature that considers all matching as 'relative choices' within the market and evaluates the economic rationality of such choices through the concept of opportunity cost.

[0233] The 7th control logic calculates the opportunity cost-based matching index through the following operation process.

[0234] The opportunity cost-based matching index is calculated by dividing the expected utility of the current choice by the market average opportunity cost.

[0235] The 8th control logic calculates the expected utility of the current choice through the following operation process.

[0236] The expected utility of the current choice is calculated by subtracting the cost of physical constraints from the product of the capability-required specification fit and the intersubjective synergy.

[0237] The ninth control logic calculates the market average opportunity cost through the following operation process.

[0238] The market average opportunity cost is calculated by adding the average value of the hypothetical expected utility with all other firms that a particular worker can choose and the average value of the hypothetical expected utility with all other workers that a particular firm can hire, and then dividing by 2.

[0239] This model goes beyond calculating the absolute value of the current choice to measure 'relative advantage' by comparing that value to the average market value of alternatives. This is a technological concept that can never be achieved through a simple combination of prior art.

[0240] Each indicator of the present invention is collected and calculated in the following specific and reproducible manner.

[0241] When worker G’s company evaluation score is 80 points, resume suitability score is 40 points, and company H’s required specification level is 35 points, the central server (100) calculates the competency-required specification suitability according to a predefined conversion formula, for example, 1 - |(company evaluation score + resume suitability score) - (company required specification level * 2)| / (maximum score sum 150), as 1 - |(80+40) - (35*2)| / 150 = 1 - 50 / 150 ≈ 0.67.

[0242] When the mutual subjective synergy is such that employee G's corporate satisfaction is 9 points and company H's employee satisfaction is 8 points, the central server (100) calculates the mutual subjective synergy as 1 + (9+8-10) / 10 = 1.7 according to the conversion formula 1 + (employee satisfaction + corporate satisfaction - 10) / 10.

[0243] Regarding the physical constraint cost, if the physical distance between worker G and company H is 50 km and the distance penalty mitigation factor is calculated to be low at 0.3, the central server (100) calculates the physical constraint cost as 50 * (1-0.3) = 35 according to the cost conversion formula physical distance * (1 - distance penalty mitigation factor).

[0244] When there are a total of 200 other companies that worker G can choose regarding the market average opportunity cost, the central server (100) uses the 8th control logic to simulate the virtual expected utility of G and each of the 200 companies and calculates that the average value is 25.5. At the same time, when there are 500 other workers that company H can hire, the average value of the virtual expected utility is calculated to be 30.2. Finally, the central server (100) calculates the market average opportunity cost as (25.5 + 30.2) / 2 = 27.85 according to the 9th control logic.

[0245] The calculated opportunity cost-based matching index is used to judge the rationality of the matching based on 1.

[0246] For a very rational choice (index > 1.2), since the current choice is more than 20% superior to other average alternatives in the market, the system immediately proposes a match to both sides.

[0247] For the possible choices (0.9 ≤ exponent ≤ 1.2), since they are reasonable choices at the market average level, they are provided in a list so that they can be compared with other candidates.

[0248] For irrational choices (index < 0.9), it is likely that better alternatives exist in the market, so the recommendation ranking of the corresponding match is lowered and other matches with higher indices are suggested preferentially.

[0249] The process of calculating the opportunity cost-based matching index using the above example is as follows.

[0250] Step 1: Calculation of current choice expected utility (8th control logic). From the value obtained by multiplying the capability-required specification fit of 0.67 and the intersubjective synergy of 1.7 (approx. 1.139), the physical constraint cost of 35 is subtracted, and the current choice expected utility is calculated as -33.861.

[0251] Step 2: Calculation of opportunity cost-based matching index (7th control logic). By dividing the current choice expected utility -33.861 by the previously calculated market average opportunity cost of 27.85, the opportunity cost-based matching index is calculated to be approximately -1.216.

[0252] Step 3: The index -1.216 calculated from the result judgment is significantly lower than the threshold value of 1, so it is judged as an 'irrational choice'.

[0253] This means that although the subjective satisfaction of the two subjects was high, the objective fit was low and the physical constraint costs were excessive, resulting in negative expected utility, and it is a very unreasonable choice compared to other alternatives in the market.

[0254] This invention is the first to introduce the economic framework of 'opportunity cost' into a matching system. This is a technical concept that transforms matching from 'isolated evaluation' to 'relative decision-making within the market.'

[0255] By calculating the final index as a ratio through division, the value of all matchings can be compared using a standardized measure of 'how superior they are compared to the market average.' While the absolute score of '80' is difficult to interpret in itself, the relative index of '1.2' provides a clear and intuitive meaning of 'a choice 20% better than the average.'

[0256] The ninth control logic calculates opportunity costs by utilizing data from the entire market accumulated in the system. This has the effect of dramatically improving the accuracy and rationality of decision-making by reflecting market trends and average value levels, which can never be known from individual user information alone, into the matching judgment.

[0257] In this specification, "optimization" has the following clear criteria.

[0258] In the present invention, "optimization" of the matching refers to the process of finally selecting and proposing a matching that has the highest value while exceeding a preset minimum threshold of 1.0 among a plurality of matching candidate groups recommended to a worker or company, based on an opportunity cost-based matching index.

[0259] Furthermore, to overcome the limitations of the prior art in evaluating the status of contract performance through static addition and subtraction of specific past events, the system of the present invention dynamically evaluates the soundness of contractual relationships by uniquely introducing a 'trust capital' model that borrows from financial asset management concepts.

[0260] This is a new approach that regards contractual relationships not as 'objects of static evaluation,' but as a dynamic process in which intangible capital called 'trust' is continuously accumulated or extinguished through interaction.

[0261] The trust capital model of the present invention calculates a comprehensive contract grade based on trust capital through the 10th control logic, the 11th control logic, and the 12th control logic.

[0262] The trust capital contract rating is a dynamic indicator representing the total amount of trust capital accumulated between workers and the company at a specific evaluation point in time.

[0263] Previous point-in-time trust capital is the trust capital contract rating value calculated in the previous evaluation cycle (e.g., the previous month), and serves as the basis for the current trust capital.

[0264] Basic trust capital is the amount of initial trust capital granted based on the soundness of the contract at the initial point in time when the contract is concluded.

[0265] The trust fluctuation trend is the amount of increase or decrease in trust that changes due to positive or negative events that occurred during the current evaluation cycle.

[0266] The positive fulfillment event value is the positive value assigned to each event that 'accumulates' trust, such as on-time wage payment, compliance with working hours, and payment of performance bonuses.

[0267] The negative performance event value is the negative value assigned to each event that 'extinguishes' trust, such as wage delays, non-contractual work instructions, and unfair treatment.

[0268] The trust depreciation rate is the rate at which trust capital naturally decreases over time in the absence of special interactions, reflecting the need for continuous effort to maintain trust relationships.

[0269] Prior art merely treats past wrongdoings (e.g., unpaid wages) as accumulated 'penalties,' failing to reflect the asymmetry of reality where it is extremely difficult to restore trust once it is broken.

[0270] This invention introduces the concept of trust capital to model these 'trust asymmetry' and 'accumulation'.

[0271] The 10th control logic calculates the current trust capital contract rating through the following calculation process.

[0272] The trust capital contract rating is calculated by adding the current cycle's trust increase / decrease flow to the value reduced by applying the trust depreciation rate to the trust capital at the previous point in time.

[0273] The 11th control logic calculates the basic trust capital at the time of contract execution through the following calculation process.

[0274] Basic trust capital is calculated by multiplying the ratio obtained by dividing the square root of the quality indicator of the employment contract by the square root of the maximum value (100) by 50, and then adding 50. This means that the better the quality of the contract, the closer the initial trust is to 100.

[0275] The 12th control logic produces a confidence increase / decrease flow through the following operation process.

[0276] The confidence increase / decrease flow is calculated by subtracting the sum of the squares of all negative fulfillment event values ​​from the sum of all positive fulfillment event values ​​that occurred during the current cycle.

[0277] The structure of the 10th control logic implies that the contract rating is a 'living indicator' that continuously changes according to current events based on past confidence capital, which is fundamentally different from prior art that evaluates only snapshots at a specific point in time.

[0278] The application of a square operation to negative event values ​​in the 12th control logic is the main structure of this model. This implies that while positive behaviors build trust linearly, negative behaviors destroy trust exponentially. This mathematically implements the real-world principle that "trust is difficult to build but can be destroyed in an instant," and possesses a depth incomparable to the simple addition and subtraction models of the prior art.

[0279] The application of a trust depreciation rate indicates that trust gradually erodes even in the best relationships without communication and positive interaction, and it is a dynamic mechanism that encourages efforts to maintain contractual relationships.

[0280] Each event value of the present invention is automatically generated based on the following specific and objective data.

[0281] Regarding the value of positive fulfillment events, +5 points are automatically awarded when the central server (100) confirms the salary transfer within the contract period for timely wage payment.

[0282] +2 points are automatically awarded if it is confirmed that there is no overtime work during the week regarding compliance with working hours.

[0283] If details of a bonus payment other than the contracted salary are confirmed regarding the payment of a performance bonus, bonus amount / 100,000 won (maximum 10 points) is automatically awarded in proportion to the bonus amount.

[0284] Regarding the negative fulfillment event value, the central server (100) detects the number of days the wage is delayed for wage delay and stores the number of days the wage is delayed itself as the negative event value. For example, if the delay is 5 days, 5 is stored.

[0285] Regarding work instructions outside the contract, a complaint is received through the worker terminal (300) stating "night work instructions not included in the contract," and if it is confirmed to be true after review by a manager, 10 points are awarded.

[0286] The calculated trust capital contract rating is used to diagnose contractual relationships in real time based on clear criteria as follows.

[0287] For excellent trust (grade ≥ 70), the company is classified as a 'Trusted Partner' and receives top exposure and recommendation bonus points within the system.

[0288] In the case of moderate reliability (40 ≤ grade < 70), it indicates that a stable relationship is maintained.

[0289] In the case of trust level (20 ≤ rating < 40), both parties are notified of a recent negative event (e.g., a 2-day wage delay) along with a warning message that 'trust capital is decreasing'.

[0290] In the case of trust bankruptcy (grade < 20), the system immediately sends a 'risk of contractual relationship bankruptcy' notification to the administrator and suggests an automated consultation service with a labor expert.

[0291] The process of explaining, in chronological order through the case of Company D, how the control logic of the present invention dynamically tracks the reliability of contractual relationships over time is as follows.

[0292] First, it is assumed that the qualitative indicator of the employment contract submitted by Company D at the initial stage of contract signing was calculated to be 81 points.

[0293] The central server (100), according to the 11th control logic, performs an operation of dividing the square root of 81 (9) by the square root of 100 (10), multiplying it by 50, and then adding 50 again, to obtain the result (9 / 10)*50 + 50 = 95. Therefore, the contractual relationship of Enterprise D starts with a high foundational trust capital of 95 points.

[0294] At the point one month after the contract, Company D generated positive events such as on-time wage payment (+5 points) and compliance with four working hours per week (+8 points), but also generated a negative event of a one-time wage delay of three days (-3 points).

[0295] At this time, the previous point in time's trust capital is 95 points, and a trust depreciation rate of 1% per month is applied. The central server (100) calculates the trust increase / decrease flow through the 12th control logic, and subtracts the squared value of negative events (3²=9) from the sum of positive events (13), so that the final trust increase / decrease flow is calculated as +4.

[0296] Next, according to the 10th control logic, the trust increase / decrease flow 4 is added to the value (94.05) of the previous trust capital of 95 points with depreciation applied, and the current trust capital after one month increases slightly to 98.05 points.

[0297] However, at the point two months later, after another month had passed, Company D had implemented on-time wage payments (+5 points) and compliance with working hours (+8 points) similar to the previous month, but caused a negative event of a severe wage delay of 10 days (-10 points).

[0298] The previous point in time's confidence capital was 98.05 points, but the 12th control logic calculates the confidence increase / decrease flow as (5+8) - (10²) = 13 - 100 = -87.

[0299] Finally, the 10th control logic reflects a massive decrease in confidence of -87 to the previous confidence capital (97.07) with depreciation applied, causing the current confidence capital after 2 months to plummet to 10.07 points.

[0300] The simulation results above clearly show that a single severe negative event can offset all the positive trust built up until then and still cause the relationship to collapse. In other words, as trust capital drops vertically from 98 points to 10 points, the system can accurately detect and warn that the contractual relationship has entered a state of 'trust bankruptcy'.

[0301] Applying a square to the negative event value in the 12th control logic to implement asymmetric trust destruction through squaring operations is a mathematical implementation of the asymmetry in reality where the speed of trust destruction caused by negative behavior is much faster and more fatal than the speed of trust accumulation brought about by positive behavior. This is a technical feature unique to the present invention that can never be captured by the symmetric addition and subtraction method of the prior art.

[0302] To model dynamic changes over time, the 10th control logic specifies that trust capital is a dynamic system that changes over time. The introduction of a trust depreciation rate implies that trust naturally decreases without continuous effort to maintain the relationship, which is an approach that makes contractual relationships subject to continuous management rather than one-time evaluation.

[0303] In this specification, "optimization" of the control logic refers to the process of adjusting the weight (positive / negative) of each event and the trust depreciation rate using 10,000 actual contract data accumulated over the past year as training data. The goal of this optimization is to make the prediction accuracy of the trust capital contract rating predicted by the system in cases where the contract was terminated within 6 months fall below the 'trust bankruptcy' threshold (20) one month before the contract is terminated 90% or higher.

[0304] The entire process of how the system of the present invention operates organically, from the matching of foreign worker 'A' and company 'B' to the evaluation of contract performance, is explained step-by-step as follows.

[0305] Step 1: Regarding information registration and data collection, Worker A inputs their resume, qualifications, desired working conditions, etc. through the Worker Terminal Unit (300) app installed on their smartphone. The person in charge of Company B also registers the job posting and required specifications through the Company Personnel Terminal Unit (400).

[0306] The data collection module (110) of the central server unit (100) receives all of the information transmitted from both terminal units, structures it, and stores it in a database.

[0307] Step 2: For the first matching and interaction, the system presents Worker A and Company B, whose basic conditions match, to each other as matching candidates based on the stored information.

[0308] A representative from Company B sees the profile of worker A and becomes interested, then sends a message requesting a video interview through the terminal unit (400).

[0309] Worker A receives a notification via the terminal unit (300) and accepts the interview request. All communication records (number of messages, interview requests) generated during this process are recorded in real time by the data collection module (110) and serve as the basis for calculating the 'distance penalty mitigation coefficient'.

[0310] Step 3: Calculation of Final Matching Index and Notification of Results. After the video interview ends, Worker A and Company B each input their subjective satisfaction with each other through their respective terminal units. Company B additionally inputs an objective evaluation score for A.

[0311] As soon as the above data is collected through the data collection module (110), the calculation module (120) of the central server unit (100) is triggered.

[0312] The calculation module (120) queries all necessary variables (evaluation score, satisfaction, distance, communication records, etc.) from the database and calculates the 'final matching index' according to the first, second, and third control logic.

[0313] The calculated index (e.g., 85.2 points) is transmitted to the result interpretation / notification module (130). The module confirms that this score exceeds the 'strong recommendation for matching' threshold (70) and generates a message saying "This is a very suitable match. Would you like to proceed with the employment contract?" and immediately sends it as a push notification to both terminal units (300, 400).

[0314] Step 4: When both parties agree to enter into a contract and create trust capital, the system provides a standard employment contract form. Company B creates and uploads the details of the contract, and the data collection module (110) receives it.

[0315] The calculation module (120) analyzes the uploaded contract to calculate the ‘employment contract qualitative indicator’ and, based on this, executes the 11th control logic to generate the ‘basic trust capital’ (e.g., 95 points) between the two parties and records it in the database.

[0316] Step 5: Regarding contract fulfillment and dynamic management of trust capital, on the monthly payday, the central server unit (100) automatically checks whether payment is made on time by linking with the payroll transfer system of Enterprise B. When payment is confirmed to be made on time, the calculation module (120) assigns a positive event value according to the 12th control logic to slightly increase the trust capital.

[0317] If an event occurs where wages are delayed by 10 days in the next month, the calculation module (120) applies a square operation to the value of the negative event to exponentially decrease the confidence capital.

[0318] When the lowered trust capital value falls below the 'trust bankruptcy' threshold (20), the result interpretation / notification module (130) immediately sends a 'high risk of labor dispute' notification to the manager terminal unit (200) to induce immediate intervention by the manager.

[0319] In this way, each component of the present invention exchanges data in real time and is organically interconnected, thereby implementing an advanced system that goes beyond simple information provision to actively intervene and manage the entire process from matching to contract execution.

[0320] Although preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above. Various modifications are possible by those skilled in the art without departing from the essence of the invention as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present invention. Explanation of the symbols

[0321] Central server (100) Data collection module (110) Calculation module (120) Result interpretation / notification module (130) Management UI module (140) Administrator terminal (200) Worker terminal (300) Company representative terminal (400)

Claims

Claim 1 A system for supporting matching between foreign workers and companies comprises: a worker terminal unit (300) receiving worker data including worker history information and mutual satisfaction; a company representative terminal unit (400) receiving company data including company requirement specification information and mutual satisfaction; a data collection module (110) connected via a network to the worker terminal unit and the company representative terminal unit, receiving the worker data and the company data and storing them in a database; and a calculation module (120) that calculates a final matching index by executing a first control logic, a second control logic, and a third control logic based on the data stored in the database. A system for supporting visa change and stay for foreign workers tailored to foreign workers, comprising: a central server unit (100) including a result interpretation / notification module (130) that determines whether a match is suitable by comparing the calculated final matching index with a preset threshold value, and then transmits the determination result to the worker terminal unit and the company representative terminal unit; wherein the first control logic calculates the final matching index by performing a multiplication operation on the objective suitability index calculated by the second control logic and the subjective matching stability index calculated by the third control logic. Claim 2 A system for supporting visa change and stay for foreign workers customized for claim 1, wherein the second control logic performs: a first operation of taking the natural logarithm of the value obtained by adding 1 to the square of the corporate evaluation score in which the company evaluated the worker; a second operation of taking the square root of the value obtained by adding 1 to the square of the difference between the worker's resume suitability score and the company's required specification level; and a third operation of adding 1 to the negative exponential function value of the value obtained by multiplying the physical distance between the worker and the company by a distance penalty mitigation coefficient calculated based on the frequency of mutual communication between the worker and the company; and calculates the objective suitability index by dividing the sum of the result of the first operation and the result of the second operation by the result of the third operation. Claim 3 A foreign worker customized visa change and stay support system according to claim 1, wherein the third control logic calculates the difference between the worker's satisfaction with the company assigned by the worker to the company and the worker's satisfaction with the company assigned by the company to the worker, multiplies the calculated difference value by a satisfaction deviation sensitivity coefficient, and then calculates the subjective matching stability index through a cosine function operation on the result. Claim 4 In claim 3, the data collection module (110) receives worker data including the worker's corporate satisfaction from the worker terminal unit (300), receives corporate data including the corporate evaluation score and the worker's satisfaction from the corporate representative terminal unit (400), transmits the address information of the worker terminal unit and the corporate representative terminal unit to a map API to receive the physical distance, and records the number of message exchanges and interview requests between the worker terminal unit and the corporate representative terminal unit to calculate the distance penalty mitigation coefficient, characterized by a customized visa change and stay support system for foreign workers. Claim 5 A system for supporting visa change and stay tailored for foreign workers according to claim 1, wherein the result interpretation / notification module (130) determines that if the final matching index is greater than or equal to a first threshold, it is 'strongly recommended for matching' and sends a notification proposing a non-face-to-face interview schedule to both terminal units; if the final matching index is less than the first threshold and greater than or equal to a second threshold, it determines that it is 'target for matching' and provides additional mutual profile information to both terminal units; and if the final matching index is less than the second threshold, it determines that it is 'unsuitable for matching' and lowers the mutual exposure ranking in the search results of both parties. Claim 6 A foreign worker customized visa change and stay support system, further comprising: an administrator terminal unit (200) for adjusting the operation policy of the system; and a management UI module (140) for providing a graphical user interface to the administrator terminal unit and changing the scaling coefficient of the first control logic, the satisfaction deviation sensitivity coefficient of the third control logic, or the threshold value of the result interpretation / notification module (130) according to input from the administrator terminal unit.

Images