Data risk identification method, system, device and product for flexible workforce

Abstract

This application discloses a method, system, device, and product for data risk identification in flexible employment. The method includes: receiving a risk verification instruction input by a user; and performing the following processing through a data risk identification intelligent agent: retrieving personnel information and four-stream data from multiple business systems based on the identifier of the flexible employment project and the risk screening time range; performing multi-dimensional risk screening on the personnel information and four-stream data to identify preliminary risk personnel with anomalies in compliance, income, workload, and / or behavior frequency; using the personnel identifier and task identifier of the preliminary risk personnel as association keys, extracting data records matching the association keys from the four-stream data for four-stream penetration association verification to verify whether the preliminary risk personnel are indeed risky personnel; and generating a risk report based on the personnel information, multi-dimensional risk screening results, and four-stream penetration association verification results. This application can accurately identify risky personnel in flexible employment and improve risk control efficiency.

Description

Technical Field

[0001] This application relates to the field of flexible employment technology, and in particular to a data risk identification method, system, electronic device, computer-readable storage medium, and computer program product for flexible employment. Background Technology

[0002] In the context of the digital economy, flexible employment models have been widely applied in various fields such as services, logistics, and technology outsourcing due to their advantages of flexible staffing and controllable costs. Flexible employment platforms typically need to connect with a massive number of employees and relatively complex employment processes, resulting in business data characterized by large personnel scale, complex cross-system flows, and rapid dynamic changes. Problems arising from various risky individuals are becoming more diversified and concealed, such as frequent occurrences of risk scenarios like working beyond age limits, abnormally high or average income, and normal income despite low workloads, posing a significant challenge to the platform's risk control. Summary of the Invention

[0003] In view of this, embodiments of this application provide a data risk identification method, system, electronic device, computer-readable storage medium, and computer program product for flexible employment, to solve at least one of the above-mentioned technical problems.

[0004] In a first aspect, embodiments of this application provide a data risk identification method for flexible employment, comprising: receiving a risk verification instruction input by a user, the risk verification instruction including the identifier of the flexible employment project and the risk screening time range; performing the following processing through a data risk identification intelligent agent: based on the identifier of the flexible employment project and the risk screening time range, retrieving personnel information and four-flow data related to the flexible employment project from multiple business systems, the four-flow data including contract flow data, business flow data, capital flow data, and invoice flow data; performing multi-dimensional risk screening on the personnel information and four-flow data based on a preset risk rule set, identifying personnel with abnormalities in the dimensions of compliance, income, workload, and / or behavior frequency, and marking them as preliminary risk personnel; for preliminary risk personnel, using the personnel identifier of the preliminary risk personnel and the task identifier completed by them within the risk screening time range as the association key, extracting data records matching the association key from the four-flow data, and performing four-flow penetration association verification on the extracted data records to verify whether the preliminary risk personnel are risky personnel; generating a risk report containing risk personnel information based on the personnel information, the multi-dimensional risk screening results, and the four-flow penetration association verification results.

[0005] According to some embodiments of this application, optionally, a four-stream pass-through correlation verification is performed on the extracted data records, including: Based on personnel and task identifiers, the contracting entity information and validity period information in the contract flow data are correlated and verified with the task recipient information and task dispatch time in the business flow data. The verification is performed to check whether the task recipient has a valid contract and whether the task scope is consistent, and to check whether the task dispatch time is within the contract validity period. Based on the task identifier, the task acceptance results and acceptance time in the business flow data are correlated and verified with the corresponding settlement amount, payment time and payee information in the fund flow data. The verification is performed to check whether the settlement amount matches the task acceptance results, whether the payment time is later than the task acceptance time, and whether the payee identity is consistent. Based on personnel and task identifiers, the signing records in the contract flow data, the task process records in the business flow data, the payment records in the fund flow data, and the invoice records in the invoice flow data are sequentially linked to construct a full-process data link from contract signing to invoice issuance, and to identify whether there are any missing or unconnectable data gaps in the full-process data link.

[0006] Optionally, according to some embodiments of this application, verifying whether the preliminary risk personnel are risk personnel with potential risks includes: determining that the preliminary risk personnel are risk personnel with contract compliance risks when the task recipient does not have a valid contract and / or the task scope is inconsistent, or when the task dispatch time is not within the contract validity period; determining that the preliminary risk personnel are risk personnel with business authenticity risks or abnormal fund entity risks when the settlement amount does not match the task acceptance results, or the payment time is earlier than the task acceptance time, or the identity of the payee is inconsistent; and determining that the preliminary risk personnel are risk personnel with process integrity risks when there are missing data or unconnectable data gaps in the entire process data link.

[0007] According to some embodiments of this application, optionally, multi-dimensional risk screening is performed on personnel information and four-flow data based on a preset risk rule set to identify personnel who are abnormal in the dimensions of compliance, income, workload and / or behavior frequency, including: calculating the unit task income of personnel based on business flow data and cash flow data; wherein, the unit task income is the ratio of settlement income to the amount of completed tasks; comparing the unit task income with a preset benchmark range of unit task income within the project; if the personnel's unit task income deviates from the benchmark range, or the personnel's workload is zero but the settlement income is not zero, then it is determined that the personnel are abnormal in the dimension of workload.

[0008] According to some embodiments of this application, optionally, multi-dimensional risk screening is performed on personnel information and four-flow data based on a preset risk rule set to identify personnel who are abnormal in the dimensions of compliance, income, workload and / or behavior frequency, including: based on personnel information, verifying whether the personnel's age is within a preset legal employment age range, and / or verifying whether the personnel's job qualifications are valid and within the certification period; based on contract flow data and fund flow data, verifying whether the personnel have settlement records in an unsigned state or continuous settlement records after the contract expires; if any of the above verification conditions are met, it is determined that the personnel are abnormal in the compliance dimension.

[0009] According to some embodiments of this application, optionally, multi-dimensional risk screening is performed on personnel information and four-flow data based on a preset risk rule set to identify personnel who are abnormal in the dimensions of compliance, income, workload and / or behavior frequency, including: calculating the personnel's income within the risk screening time range based on cash flow data; comparing the income with a preset income benchmark value within the project; if the deviation between the income and the income benchmark value within the project exceeds a preset threshold, or the income exceeds a preset static settlement amount limit, then it is determined that the personnel are abnormal in the income dimension.

[0010] According to some embodiments of this application, optionally, multi-dimensional risk screening is performed on personnel information and four-flow data based on a preset risk rule set to identify personnel who are abnormal in the dimensions of compliance, income, workload and / or behavior frequency, including: based on business flow data and capital flow data, calculating the settlement frequency and / or service area switching frequency of personnel within a preset time window; comparing the settlement frequency and / or service area switching frequency with a preset frequency threshold; if the personnel's settlement frequency or service area switching frequency exceeds the corresponding frequency threshold, it is determined that the personnel are abnormal in the dimension of behavior frequency.

[0011] According to some embodiments of this application, optionally, before generating a risk report, the data risk identification method for flexible employment further includes: performing the following processing through a data risk identification agent: for risk personnel determined to be at risk through four-stream penetration correlation verification, assigning a basic risk score to each identified risk behavior based on the risk type; if the number of consecutive occurrences of the risk behavior in historical data exceeds a preset threshold or the risk personnel exist in the historical risk list, then adding an additional persistent risk score or recurring risk score; accumulating all risk scores of the risk personnel to obtain the total risk score of the risk personnel; and determining the risk level of the risk personnel based on a preset total score threshold range.

[0012] According to some embodiments of this application, optionally, generating a risk report containing information on at-risk personnel includes: generating a structured list of at-risk personnel, the list including information on at least one at-risk personnel, the information including at least one of personnel identification, risk level, risk type, abnormal behavior, verification basis, and data links to related evidence; generating a risk classification statistical report, used to perform quantity statistics and / or percentage analysis of identified at-risk personnel according to risk type and / or risk level; generating corresponding risk management recommendations for at-risk personnel of different risk levels or different risk types; and generating a risk report based on the list of at-risk personnel, the risk classification statistical report, and the risk management recommendations.

[0013] According to some embodiments of this application, optionally, before performing multi-dimensional risk screening of personnel information and four-stream data based on a preset risk rule set, the data risk identification method for flexible employment further includes: performing the following processing through a data risk identification intelligent agent: calculating a risk assessment benchmark value for the flexible employment project based on the four-stream data of the flexible employment project within the risk screening time range; wherein, the risk assessment benchmark value includes at least one of the benchmark range of per capita income, benchmark value of income per single task, and benchmark value of per capita task volume; generating a risk report containing information on risky personnel, further including: comparing the calculated risk assessment benchmark value with preset industry benchmark data; if the deviation between the risk assessment benchmark value and the corresponding industry benchmark data exceeds a preset deviation threshold, then marking the risk assessment benchmark value as abnormal in the risk report.

[0014] Secondly, embodiments of this application provide a data risk identification system for flexible employment, comprising: an interaction module for receiving risk verification instructions input by a user, the risk verification instructions including the identifier of the flexible employment project and the risk screening time range; and a data risk identification intelligent agent for performing the following processes: retrieving personnel information and four-flow data related to the flexible employment project from multiple business systems based on the identifier of the flexible employment project and the risk screening time range, the four-flow data including contract flow data, business flow data, fund flow data, and invoice flow data; and performing data analysis on the personnel information and the four-flow data based on a preset risk rule set. The data undergoes multi-dimensional risk screening to identify individuals exhibiting anomalies in compliance, income, workload, and / or behavioral frequency, marking them as preliminary risk personnel. For these preliminary risk personnel, data records matching the association key are extracted from the four-stream data using their personnel identifier and the task identifiers they completed within the risk screening timeframe. A four-stream penetration-based association verification is then performed on the extracted data records to confirm whether the preliminary risk personnel are indeed at risk. Based on the personnel information, the multi-dimensional risk screening results, and the four-stream penetration-based association verification results, a risk report containing information about the risk personnel is generated.

[0015] Thirdly, embodiments of this application provide an electronic device, which includes a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the steps of the data risk identification method for flexible employment as described above.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the steps of the data risk identification method for flexible employment as described above.

[0017] Fifthly, embodiments of this application provide a computer program product, which includes computer program instructions that, when executed by a processor, implement the steps of the data risk identification method for flexible employment as described above.

[0018] The data risk identification method, system, electronic device, computer-readable storage medium, and computer program product for flexible employment, as described in the embodiments of this application, achieve the following: First, by using a data risk identification intelligent agent based on the identifier of the flexible employment project and the screening time range, it can automatically retrieve personnel information and data from four flows: contract flow, business flow, capital flow, and invoice flow, without manual intervention. This achieves automated linkage retrieval and integration of data from multiple systems, solving the problems of data dispersion, low collaboration efficiency, and high labor costs in the traditional model, and ensuring the timeliness and completeness of data retrieval. Second, based on the autonomous processing capability of the data risk identification intelligent agent, it can autonomously identify preliminary risk personnel based on compliance, income, workload, and behavior frequency through multi-dimensional risk screening. Then, it automatically performs a four-flow data penetration-style correlation verification using personnel identifiers and task identifiers as association keys to verify the business authenticity, data consistency, and risk correlation of preliminary risk personnel. This can accurately identify risk personnel who exceed age limits, have abnormally deviated income from the average, or have low workloads but high settlement amounts. Compared with manual screening and a single rule engine, it can reduce the false positive rate and false negative rate, and improve the accuracy and intelligence level of risk personnel identification. On the other hand, from data retrieval, multi-dimensional screening, in-depth verification to risk report generation, the entire process is completed autonomously by the data risk identification intelligent agent, realizing standardized closed-loop processing of the entire risk identification process. This not only adapts to the weekly and monthly routine risk control needs of flexible employment, compressing the traditional manual screening work that takes several days to complete in a short time, thus improving the efficiency of risk identification, but also reduces the reliance on human experience, while providing complete data support for subsequent risk review and rule optimization. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings of the embodiments of this application will be briefly described below.

[0020] Figure 1 This is a flowchart illustrating a data risk identification method for flexible employment provided in an embodiment of this application.

[0021] Figure 2 This is a schematic flowchart of step S1022 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0022] Figure 3 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0023] Figure 4 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0024] Figure 5 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0025] Figure 6 This is a flowchart illustrating step S1023 of the data risk identification method for flexible employment provided in an embodiment of this application.

[0026] Figure 7 This is another schematic diagram of S1023 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0027] Figure 8 This is a schematic flowchart of S1024 in the data risk identification method for flexible employment provided in the embodiments of this application.

[0028] Figure 9 This is a structural block diagram of a data risk identification system for flexible employment provided in an embodiment of this application.

[0029] Figure 10 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0030] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0032] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0033] Various modifications and variations can be made to this application without departing from its spirit or scope, which will be apparent to those skilled in the art. Therefore, this application is intended to cover modifications and variations falling within the scope of the corresponding claims (the claimed technical solutions) and their equivalents. It should be noted that the implementation methods provided in the embodiments of this application can be combined with each other without contradiction.

[0034] Before describing the technical solutions provided in the embodiments of this application, in order to facilitate understanding of the embodiments of this application, this application first specifically explains the problems existing in the prior art: In the context of the digital economy, flexible employment models have been widely applied in various fields such as services, logistics, and technology outsourcing due to their advantages of flexible staffing and controllable costs. Flexible employment platforms typically need to connect with a massive number of employees and relatively complex employment processes, resulting in business data characterized by large personnel scale, complex cross-system flows, and rapid dynamic changes. Problems arising from various risky individuals are becoming more diversified and concealed, such as frequent occurrences of risk scenarios like working beyond age limits, abnormally high or average income, and normal income despite low workloads, posing a significant challenge to the platform's risk control.

[0035] In the context of the digital economy, flexible employment models, due to their advantages of flexible staffing and controllable costs, have been widely applied in various fields such as services, logistics, and technology outsourcing, becoming an important choice for enterprises to optimize their employment structure and improve operational efficiency. Against this backdrop, flexible employment platforms, as the core hub connecting employers with a massive number of flexible workers, typically need to simultaneously connect with thousands or even tens of thousands of employees, handling all aspects of the employment process, including task assignment, contract signing, results acceptance, and settlement payments. The business data of flexible employment platforms not only exhibits characteristics of a large personnel scale, complex cross-business system flows, and rapid real-time dynamic changes, but also covers multi-dimensional content such as basic personnel information, task performance data, and fund settlement records. At the same time, compliance and operational issues caused by various risky personnel are becoming increasingly prominent, showing a diversified and concealed development trend, such as fraudulent employment, misappropriation of funds, and violations, posing significant challenges to the risk control work of flexible employment platforms.

[0036] To address at least one of the aforementioned technical problems, this application provides a data risk identification method, system, electronic device, computer-readable storage medium, and computer program product for flexible employment.

[0037] The following section first introduces the data risk identification method for flexible employment provided in this application.

[0038] The data risk identification method for flexible employment provided in this application embodiment can be mainly implemented through a data risk identification intelligent agent. The data risk identification intelligent agent is set up with a workflow for identifying risky personnel in flexible employment scenarios. Based on models, algorithms and / or knowledge bases fine-tuned for flexible employment scenarios, it can be mainly used to perform multi-dimensional risk screening and four-flow penetration correlation verification of personnel information and four-flow data to accurately identify risky personnel in flexible employment scenarios.

[0039] Figure 1 This is a flowchart illustrating a data risk identification method for flexible employment provided in an embodiment of this application. Figure 1 As shown, the data risk identification method for flexible employment provided in this application embodiment may include the following steps S101, S1021 to S1024.

[0040] S101: Receive risk verification instructions from users. The risk verification instructions include the identifier of the flexible employment project and the time range for risk screening.

[0041] Specifically, users can input risk verification instructions (i.e., risk verification requests) through the risk control interface of the flexible employment platform. The risk verification instructions can include the identifier of the flexible employment project and the time range for risk screening. The identifier of the flexible employment project can include the project name and / or ID, used to locate the flexible employment project to be verified. The time range for risk screening can be any period within the entire cycle of the flexible employment project; it can cover the entire project cycle, or select a single day, week, month, quarter, or custom period within the project. The specific time range can be flexibly adjusted according to actual circumstances, and this application does not impose any limitations on it.

[0042] In some embodiments, the risk verification instruction may optionally include key monitoring personnel, such as high-income personnel, newly registered personnel, cross-regional service personnel, or personnel holding relevant certifications. This allows the data risk identification agent to focus on key monitoring personnel, narrowing the screening scope, improving the targeting and efficiency of risk identification, and reducing resource waste caused by indiscriminate screening. In some embodiments, users can also customize risk thresholds through the risk control operation interface of the flexible employment platform, such as the monthly settlement amount limit, the percentage deviation of single-task income from the benchmark value, and the threshold for the number of high-frequency settlements, etc., which are not limited in this application.

[0043] The data risk identification agent performs the following steps S1021 to S1024.

[0044] S1021: Based on the identification of flexible employment projects and the time range of risk investigation, retrieve personnel information and four-flow data related to flexible employment projects from multiple business systems. The four-flow data includes contract flow data, business flow data, capital flow data and invoice flow data.

[0045] The data risk identification intelligent agent can interface with multiple business systems, including Customer Relationship Management (CRM), Flexible Employment Platform (FEP), Big Data Service (BDS), and personnel basic information database. Based on the identification of flexible employment projects and the time frame for risk assessment, the data risk identification intelligent agent can retrieve personnel information and four-stream data related to flexible employment projects from multiple business systems through interfaces.

[0046] The personnel information may include personnel authentication information, age, qualifications, and / or registration time, etc. The four-flow data may include contract flow data, business flow data, fund flow data, and invoice flow data. For example, contract flow data may include project cooperation contracts, personnel service agreements, and other terms and performance records; business flow data may include task assignment vouchers, execution vouchers, acceptance vouchers, and completion vouchers; fund flow data may include service fee settlements, salary payments, and bank transfer records; and invoice flow data may include invoice information, invoice authentication, and tax item matching data.

[0047] S1022: Based on a preset set of risk rules, conduct multi-dimensional risk screening of personnel information and four-stream data to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, and mark them as personnel with preliminary risks.

[0048] The data risk identification intelligent agent can perform multi-dimensional risk screening on personnel information and four-stream data using a pre-set set of risk rules. These multi-dimensional aspects can include at least two of the following: compliance, revenue, workload, and behavior frequency.

[0049] In terms of compliance, the system can verify whether employees' ages are within the legal employment range, whether their qualifications are valid, and whether their contracts are valid, thus identifying individuals with abnormalities such as those exceeding age limits, expired qualifications, or settlements without contracts. In terms of income, it can compare the average and median income within a project with a set threshold to identify individuals whose income deviates from the benchmark or exceeds the threshold. In terms of workload, it can verify the number of tasks undertaken and completed by employees within the risk assessment period and correlate this with income data to determine the reasonableness of the workload. In terms of behavioral frequency, it can identify individuals exhibiting abnormal behaviors such as high-frequency settlements or high-frequency order taking across regions. The data risk identification intelligence agent can uniformly mark abnormal individuals identified from each dimension, generating a preliminary list of risky personnel. This preliminary list can also indicate the abnormal dimension and reason for each individual's abnormality.

[0050] S1023: For personnel with preliminary risk, the personnel identifier of the personnel with preliminary risk and the task identifier of the task completed within the risk investigation time range are used as the association key. Data records that match the association key are extracted from the four-stream data, and the extracted data records are subjected to four-stream penetration association verification to verify whether the personnel with preliminary risk are risky personnel.

[0051] For each individual initially identified as a risk, the data risk identification agent can use their personnel identifier and the task identifiers they completed within the risk assessment period as dual key associations to extract corresponding data records from the retrieved four-flow data. These records include contract signing information, task assignment vouchers, acceptance vouchers, settlement payment details, and invoices. Then, a four-flow penetration-based correlation verification is performed on the extracted data records. This verification checks the consistency between contract flow and business flow, the consistency between business flow and cash flow, and whether there are any missing or disconnected data points in the entire data chain. If the four-flow penetration-based correlation verification passes, the individual is excluded from risk suspicion and the initial risk label is removed. If the four-flow penetration-based correlation verification fails, the individual is determined to be a risky individual with inherent risks.

[0052] S1024: Generate a risk report containing information on at-risk personnel based on personnel information, multi-dimensional risk screening results, and four-flow penetration-based correlation verification results.

[0053] The data risk identification intelligent agent integrates personnel information, multi-dimensional risk screening results, and four-flow penetration-based correlation verification results to generate a risk report containing information on at-risk personnel.

[0054] The risk report should include at least the basic information of the risk personnel, multi-dimensional risk screening results, and four-flow penetration verification results. The basic information of the risk personnel can include their identification, name, and qualifications. Furthermore, the risk report can also include four-flow data vouchers associated with the risk personnel, such as screenshots of abnormal settlement records and scanned copies of contracts, enabling risk traceability. Finally, the data risk identification intelligent agent can push the generated risk report to the user's risk control interface, allowing users to view details and export files, thus providing precise guidance for offline verification and risk handling.

[0055] The data risk identification method for flexible employment embodiments of this application, on the one hand, automatically retrieves personnel information and data from four flows—contract flow, business flow, capital flow, and invoice flow—based on the identification of flexible employment projects and the screening time range, without manual intervention. This achieves automated linkage retrieval and integration of data from multiple systems, solving the problems of data dispersion, low collaboration efficiency, and high labor costs in the traditional model, and ensuring the timeliness and completeness of data retrieval. On the other hand, based on the autonomous processing capabilities of the data risk identification intelligent agent, it autonomously identifies preliminary risk personnel based on compliance, income, workload, and behavior frequency through multi-dimensional risk screening. Then, it automatically performs a four-flow data penetration-style correlation verification using personnel identifiers and task identifiers as association keys to verify the business authenticity, data consistency, and risk correlation of preliminary risk personnel. This can accurately identify risk personnel who exceed age limits, have abnormally deviated income from the average, or have low workloads but high settlement amounts. Compared with manual screening and a single rule engine, this can reduce the false positive rate and false negative rate, and improve the accuracy and intelligence level of risk personnel identification. On the other hand, from data retrieval, multi-dimensional screening, in-depth verification to risk report generation, the entire process is completed autonomously by the data risk identification intelligent agent, realizing standardized closed-loop processing of the entire risk identification process. This not only adapts to the weekly and monthly routine risk control needs of flexible employment, compressing the traditional manual screening work that takes several days to complete in a short time, thus improving the efficiency of risk identification, but also reduces the reliance on human experience, while providing complete data support for subsequent risk review and rule optimization.

[0056] To facilitate understanding, the following examples illustrate the data risk identification method for flexible employment.

[0057] Figure 2 This is a schematic flowchart of step S1022 in the data risk identification method for flexible employment provided in an embodiment of this application. Figure 2 As shown, according to some embodiments of this application, optionally, S1022: performing multi-dimensional risk screening on personnel information and four-stream data based on a preset risk rule set to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, may include the following steps S201 to S203.

[0058] S201: Based on personnel information, verify whether the personnel's age is within the preset legal working age range, and / or verify whether the personnel's job qualifications are valid and within the certification period.

[0059] In S201, the data risk identification agent can extract information such as age and / or professional qualification certificates from personnel information. Then, it verifies whether the personnel's age falls within a preset legal working age range, such as 18 to 65 years old. The specific legal working age range can be fine-tuned according to the specific job requirements; this application does not limit this. And / or, it verifies whether the personnel possess professional qualification certificates, whether the certificates are within their certification period, and whether the certification body is valid, screening out personnel with uncertified qualifications, expired certificates, or forged qualifications, and simultaneously recording specific information for each type of anomaly, such as a professional qualification certificate expiring in March 2025.

[0060] S202: Based on contract flow data and cash flow data, verify whether personnel have settlement records in an unsigned state or continuous settlement records after the contract expires.

[0061] In S202, the data risk identification agent can verify whether personnel have any valid signed contracts within the risk screening timeframe, meaning the contracts are in effect and within their validity period. If there is no corresponding signing record in the contract flow data, or the signing record is ineffective or terminated, but there are concurrent settlement payment records in the cash flow data, it is determined to be a settlement anomaly in the unsigned state. Additionally, the contract validity period in the contract flow data can be compared with the settlement records in the cash flow data. If a preset number of settlement records occur after the contract validity period expires and there are no renewal records, it is determined to be a continuous settlement anomaly after the contract expires.

[0062] S203: If any of the above verification conditions are met, the person is deemed to have an anomaly in the compliance dimension.

[0063] The data risk identification intelligent agent summarizes the verification results of S201 and S202. If a person has any of the four situations, such as exceeding the age limit, invalid qualifications, no contract settlement, or expired contract settlement, the person will be automatically marked as an abnormal person in the compliance dimension, included in the preliminary risk personnel list, and the abnormal type will be clearly marked, such as exceeding the age limit or no contract settlement.

[0064] Thus, by using data-driven risk identification agents based on dimensions such as age, qualifications, and contract signing and settlement, it is possible to identify illegal employment situations such as exceeding age limits and invalid qualifications, reducing compliance risks such as legal disputes and tax penalties faced by the platform due to illegal employment. Furthermore, through intelligent linkage verification of contract flow and fund flow, it is possible to accurately identify fraudulent employment situations such as settlement without a contract or settlement after the contract expires, which helps to address operational risks such as misappropriation of funds and fraudulent employment arising from the disconnect between contracts and settlements in flexible employment scenarios.

[0065] Figure 3 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application. For example... Figure 3 As shown, according to some embodiments of this application, optionally, S1022: performing multi-dimensional risk screening on personnel information and four-stream data based on a preset risk rule set to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, may include the following steps S301 to S303.

[0066] S301: Based on cash flow data, calculate the income of personnel within the risk screening period.

[0067] In S301, the data risk identification agent can extract all settlement records of an individual within the risk assessment period from the cash flow data, using the individual's identifier as the association key. Then, based on these settlement records, the agent calculates the individual's income within the risk assessment period. After calculation, the calculation details can be retained, making the income data traceable and verifiable.

[0068] S302: Compare the deviation of the revenue from the preset benchmark revenue value within the project.

[0069] In S302, the data risk identification agent can compare the deviation of an individual's income (hereinafter referred to as actual income) within the risk screening timeframe with a preset project income benchmark. For example, based on a mean or median deviation calculation algorithm, the degree of deviation between the individual's actual income and the project income benchmark can be quantified to obtain a deviation value. For instance, if an individual's actual income is 2.8 times higher than the project income benchmark, the deviation reaches 180%. For example, the project income benchmark can be calculated based on historical project data from the BDS system, specifically the average income per person within the project or the median income within the project.

[0070] S303: If the deviation between income and the income benchmark value within the project exceeds a preset threshold, or if the income exceeds the preset static settlement amount limit, then the person is deemed to have an abnormality in the income dimension.

[0071] If an employee's income deviates from the project's benchmark income by more than a preset threshold (e.g., ±3 times), regardless of whether the income is too high or too low, it is marked as an abnormal income deviation. Additionally, if an employee's income exceeds a preset static settlement amount limit, it is marked as a high-income exceeding-limit abnormality. The preset threshold and static settlement amount limit can be flexibly adjusted according to actual circumstances; this application does not impose any restrictions on them.

[0072] If any of the conditions of abnormal income deviation or abnormal high income exceedance are met, the person is determined to have an abnormality in the income dimension, is included in the preliminary risk personnel list, and the abnormality type and judgment basis are marked.

[0073] Thus, addressing the characteristics of large income fluctuations and significant differences in job-specific income among flexible workers, the data risk identification intelligent agent, using a project-specific income benchmark as a reference and combined with a quantitative deviation algorithm, can accurately identify risk situations such as income deviating from a reasonable range or exceeding the settlement limit. This solves the problems of a lack of unified judgment standards for income anomalies and the easy omission of hidden high or low income anomalies, reducing the risks of fund loss and tax violations caused by the mismatch between income and actual labor in flexible employment scenarios. Furthermore, the data risk identification intelligent agent automatically completes the entire process of income aggregation, benchmark comparison, and deviation calculation, reducing manual accounting costs and helping to improve the efficiency and accuracy of income anomaly screening, meeting the risk control needs of flexible employment platforms for income accounting of massive numbers of personnel.

[0074] Figure 4 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application. For example... Figure 4 As shown, according to some embodiments of this application, optionally, S1022: performing multi-dimensional risk screening on personnel information and four-flow data based on a preset risk rule set to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, may include the following steps S401 to S403.

[0075] S401: Calculate the unit task income of personnel based on business flow data and cash flow data; where unit task income is the ratio of settlement income to the amount of completed tasks.

[0076] In S401, the data risk identification agent can use personnel identifiers as the association key to link and match personnel's business flow data and cash flow data within the risk assessment timeframe. It extracts the number of valid tasks completed and accepted by the personnel within the risk assessment timeframe from the business flow data, and extracts the actual settlement amount matching the valid tasks from the cash flow data. Then, based on the actual settlement amount and the number of valid tasks, it calculates the personnel's unit task income, which is the quotient of the actual settlement amount and the number of valid tasks. After the calculation is completed, it retains the valid tasks, their quantities, and the actual settlement amount used in the calculation, making the unit task income data traceable and verifiable.

[0077] S402: Compare the unit task revenue with the preset benchmark range of unit task revenue within the project.

[0078] In S402, the data risk identification agent can compare the unit task revenue of personnel within the risk assessment timeframe with a preset benchmark range for unit task revenue within the project. For example, the benchmark range for unit task revenue within the project can be calculated based on historical operational data from the BDS system, specifically a reasonable fluctuation range, such as ±50%, of the mean or median of unit task revenue within the project.

[0079] S403: If a person's unit task income deviates from the benchmark range, or if a person's task volume is zero but the settlement income is not zero, then the person is deemed to have an anomaly in the task volume dimension.

[0080] If an employee's unit task income deviates from the preset benchmark range for unit task income within the project (i.e., unit task income is higher than the upper limit or lower than the lower limit of the benchmark), it can be marked as an anomaly in unit task income matching. Additionally, if an employee has zero valid tasks, but their settlement income for the corresponding period is not zero (i.e., no tasks but settlement), it is marked as an extreme anomaly in task income. If either the unit task income matching anomaly or the extreme anomaly in task income is met, the employee is determined to have an anomaly in the task volume dimension, is included in the preliminary risk personnel list, and the anomaly type and judgment basis are noted.

[0081] Thus, addressing the challenges of diverse task types and varying order volumes across flexible employment platforms, a data risk identification intelligent agent links business and financial flow data to calculate unit task revenue and compare it with a benchmark range within the project. This accurately identifies risky situations such as unit task revenue deviating from a reasonable range or settlements occurring despite no tasks. It resolves the lack of unified standards for task-revenue matching verification and the ease of overlooking concealed anomalies, reducing risks such as fraudulent employment and fund misappropriation arising from task-revenue mismatches in flexible employment scenarios. Furthermore, the data risk identification intelligent agent automates the entire process of task volume and revenue data linkage matching, unit task revenue calculation, and benchmark range comparison, reducing manual comparison costs and improving the efficiency and accuracy of anomaly screening in the task volume dimension. This meets the risk control needs of flexible employment platforms for verifying the task-revenue matching of massive numbers of personnel.

[0082] Figure 5 This is another schematic diagram of S1022 in the data risk identification method for flexible employment provided in the embodiments of this application. For example... Figure 5 As shown, according to some embodiments of this application, optionally, S1022: performing multi-dimensional risk screening on personnel information and four-flow data based on a preset risk rule set to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, may include the following steps S501 to S503.

[0083] S501: Based on business flow data and cash flow data, the settlement frequency and / or service area switching frequency of the statistician within a preset time window.

[0084] In S501, the data risk identification agent first retrieves a preset time window parameter. This preset time window parameter can be set to a single day, three days, or one week, depending on risk control requirements; this application does not limit this setting. Then, using the personnel identifier as the association key, it extracts valid settlement records within the preset time window from the cash flow data and calculates the quotient of the total number of settlements and the preset time window to obtain the settlement frequency per unit time, such as three settlements within one day. When calculating the service area switching frequency, service area information (such as city and district) for each task can be extracted from the business flow data. After removing duplicate areas, the number of different service areas switched within the time window and the switching frequency are calculated, such as accepting orders across more than three cities within one week, resulting in two area switches.

[0085] S502: Compare the settlement frequency and / or service area switching frequency with a preset frequency threshold.

[0086] In S502, the data risk identification agent can compare the settlement frequency and / or service area switching frequency of personnel with preset corresponding frequency thresholds. For example, the frequency thresholds can be set based on industry-standard operational data or project historical risk data, and their specific values ​​can be flexibly adjusted according to actual circumstances; this application does not limit this. For instance, in some examples, the settlement frequency threshold can be set to ≤2 settlements per day or ≤5 settlements per week, and the service area switching frequency threshold can be set to ≤2 cross-regional changes per week or ≤1 regional change within 3 days; this application does not limit this.

[0087] S503: If a person's settlement frequency or service area switching frequency exceeds the corresponding frequency threshold, it is determined that the person has an abnormality in the behavioral frequency dimension.

[0088] If a person's actual settlement frequency exceeds the preset settlement frequency threshold, it is marked as a high-frequency settlement anomaly; if a person's actual service area switching frequency exceeds the preset service area switching frequency threshold, it is marked as a cross-region switching exceeding the standard anomaly.

[0089] If either the high-frequency settlement anomaly or the cross-regional switching exceeding the limit anomaly is met, the person being assessed is considered to have an abnormal behavior frequency dimension and will be included in the preliminary risk personnel list, with the anomaly type and the basis for assessment noted. For example, the anomaly type may include abnormal behavior frequency (such as high-frequency settlement on a single day) or abnormal behavior frequency (such as exceeding the cross-regional limit within one week).

[0090] In this way, by using a data risk identification intelligent agent to identify individuals with abnormal behavior frequency, manual verification of each transaction is replaced, improving the efficiency of identifying abnormal behaviors such as high-frequency settlements and cross-regional order acceptance. Simultaneously, it supports user-defined frequency thresholds and time windows, adapting to the business characteristics of different projects and positions, thus improving the scenario adaptability of the screening. Furthermore, by filtering individuals with abnormal behavior frequency in advance, subsequent four-stream penetration verification can focus on the target, reducing the cost of ineffective verification and effectively reducing violations such as fraudulent order acceptance and risk control evasion.

[0091] Figure 6 This is a schematic flowchart of step S1023 in the data risk identification method for flexible employment provided in an embodiment of this application. Figure 6 As shown, according to some embodiments of this application, optionally, S1023: performing four-stream pass-through association verification on the extracted data records may include the following steps S601 to S603.

[0092] S601: Based on personnel identifiers and task identifiers, perform correlation verification between the contracting entity information and validity period information in the contract flow data and the task recipient information and task dispatch time in the business flow data. Verify whether the task recipient has a valid contract and whether the task scope is consistent, and verify whether the task dispatch time is within the contract validity period.

[0093] In S601, the data risk identification agent uses personnel identifiers and task identifiers as dual association keys to simultaneously retrieve corresponding contract flow data and business flow data. From the contract flow data, it extracts information on the contracting entity and validity period. Contracting entity information may include the contracting entity's name, unified social credit code or personal ID number, contracting entity type, and the scope of services stipulated in the contract. Validity period information may include the contract's effective date, contract termination date, and contract renewal records. From the business flow data, it extracts information on the task recipient and the task assignment time. Task recipient information may include the task recipient's name, personal ID number or unified social credit code, contact number, task content, and service area. Then, the data risk identification agent performs correlation verification on the contract flow data and business flow data. First, it verifies whether the task recipient has a valid contract and whether the task content falls within the scope of services stipulated in the contract, identifying situations where tasks are undertaken without a contract or beyond the scope of the contract. Second, it verifies whether the task assignment time is within the contract's validity period, identifying anomalies such as assignments after the contract expires or before the contract takes effect. Secondly, synchronously record and verify discrepancies, such as the task recipient not matching the contracting entity or the dispatch time exceeding the contract validity period.

[0094] S602: Based on the task identifier, perform correlation verification between the task acceptance results and acceptance time in the business flow data and the corresponding settlement amount, payment time, and payee information in the fund flow data. Verify whether the settlement amount matches the task acceptance results, whether the payment time is later than the task acceptance time, and whether the payee's identity is consistent.

[0095] In S602, the data risk identification agent uses the task identifier as the association key to synchronously retrieve corresponding business flow data and fund flow data. It extracts task acceptance results and acceptance time from the business flow data. Task acceptance results may include acceptance vouchers, quantitative indicators of results, and task completion confirmation information. It extracts settlement amount, payment time, and payee information from the fund flow data. Payee information may include payee name, personal ID number or unified social credit code, bank account information, and bank name. Then, the data risk identification agent performs correlation verification on the business flow data and fund flow data. First, it verifies whether the settlement amount matches the task acceptance results based on the single task revenue benchmark within the project, identifying situations where the settlement amount does not match the task value. Second, it verifies whether the actual payment time is later than the acceptance completion time, i.e., whether it conforms to the business logic of acceptance before settlement, identifying anomalies such as payment before acceptance or advance payment. Third, it verifies whether the payee information matches the task recipient information, identifying risks of collection on behalf of others or fraudulent collection. Finally, it synchronously records verification anomalies and specific data to provide a basis for subsequent risk assessment.

[0096] S603: Based on personnel and task identifiers, the contract signing records in the contract flow data, the task full-process records in the business flow data, the payment records in the fund flow data, and the invoice issuance records in the invoice flow data are sequentially linked to construct a full-process data link from contract signing to invoice issuance, and identify whether there are missing or unconnectable data gaps in the full-process data link.

[0097] In S603, the data risk identification agent uses personnel and task identifiers as dual association keys to simultaneously retrieve signing records from contract flow data, task completion records from business flow data, payment records from fund flow data, and invoice records from invoice flow data. Following the business sequence of contract signing → task assignment → task completion → task acceptance → settlement payment → invoice issuance, the signing records, task completion records, payment records, and invoice records are sequentially linked to construct a complete data link from contract signing to invoice issuance. Then, the data risk identification agent verifies the integrity of the link node by node, identifying whether there are missing key data, such as no contract signing record, no task acceptance record, or payment records without corresponding invoice records. It also identifies gaps in data association, such as mismatches between invoice and settlement amounts, discrepancies between the invoice recipient and the contract signatory, or invoice issuance time preceding payment time. Simultaneously, it analyzes the gaps and their specific causes, generating a link verification report.

[0098] In this way, by using a data risk identification intelligent agent to perform four-flow penetration and correlation verification of data, it is possible to accurately review the personnel initially identified in the multi-dimensional risk screening, verify the authenticity of the entire business process layer by layer, accurately identify fraudulent employment situations such as business without contracts, settlement without acceptance, and false invoices, and accurately locate hidden risk points and data gaps. This effectively solves the core pain points of data disconnection and difficulty in verifying the authenticity of business in flexible employment scenarios, reduces the misjudgment rate and missed judgment rate of risk assessment, reduces tax penalties and legal disputes caused by inconsistencies in the four flows, and improves the compliance operation and fund security management capabilities of flexible employment platforms.

[0099] In addition, given the complex business processes and dispersed risk points of flexible employment, the system focuses on core compliance nodes through layered verification and conducts full-chain checks covering the entire business cycle, forming a dual guarantee of point verification and chain traceability. This solves the problems of incomplete coverage by a single verification dimension and insufficient basis for risk judgment, helping the platform to curb fraudulent employment, misappropriation of funds, and illegal invoices from the source.

[0100] Figure 7 This is another schematic diagram of the process of S1023 in the data risk identification method for flexible employment provided in the embodiments of this application. Figure 7 As shown, according to some embodiments of this application, optionally, S1023: verifying whether the preliminary risk personnel are risk personnel with risks may include the following steps S701 to S703.

[0101] S701: When the task recipient does not have a valid contract and / or the task scope is inconsistent, or the task assignment time is not within the contract validity period, the preliminary risk personnel are determined to be risk personnel with contract compliance risks.

[0102] In S701, the data risk identification agent makes judgments based on the contract flow and business flow correlation verification results of S601. If the verification results show any of the following abnormalities: the task recipient does not have a valid signed contract, the task content exceeds the service scope agreed in the contract, or the task assignment time is not within the contract validity period, then the person with the initial risk is determined to be a person with risk, and the risk type is marked as contract compliance risk.

[0103] S702: When the settlement amount does not match the task acceptance results, or the payment time is earlier than the task acceptance time, or the identity of the payee is inconsistent, the preliminary risk personnel are determined to be risk personnel with business authenticity risk or abnormal fund entity risk.

[0104] In S702, the data risk identification agent makes judgments based on the business flow and fund flow correlation verification results of S602. If the verification results show any of the following anomalies: the settlement amount does not match the task acceptance result, the payment time is earlier than the task acceptance time, or the payee information is inconsistent with the task recipient information, the person with the initial risk is determined to be at risk. Simultaneously, the corresponding risk type is marked according to the specific anomaly. For example, when the settlement amount does not match the task acceptance result, or the payment time is earlier than the task acceptance time, the risk type is marked as business authenticity risk. When the payee information is inconsistent with the task recipient information, the risk type is marked as abnormal fund entity risk.

[0105] S703: When there are missing or unconnectable data gaps in the entire data chain, the personnel initially identified as having risks to the integrity of the process are identified.

[0106] In S703, the data risk identification agent makes judgments based on the results of the four-flow end-to-end data link verification in S603. If the verification results show that there is a critical data gap in the end-to-end data link, or a gap in data that cannot be linked, regardless of whether a single node is missing or multiple nodes are abnormally linked, the person with the initial risk can be identified as a person with risk, and the risk type is marked as process integrity risk.

[0107] Thus, by using a data risk identification intelligent agent to classify risks based on the results of four-flow penetration-based correlation verification, accurate final risk identification can be achieved for personnel initially identified through multi-dimensional risk screening. This ensures a one-to-one correspondence between risk types and abnormal situations, thereby solving the problems of ambiguous risk assessment and lack of clear risk type classification in flexible employment risk control. Furthermore, clear risk type classification facilitates targeted risk control measures on the platform, improving the accuracy and efficiency of risk handling. This helps curb behaviors such as fraudulent employment, misappropriation of funds, and invoice violations caused by different types of risks, enhancing the overall compliance and risk control capabilities of the flexible employment platform.

[0108] According to some embodiments of this application, optionally, before S1024: generating a risk report, the data risk identification method for flexible employment may further include performing steps one to three by a data risk identification agent.

[0109] Step 1: For risk personnel identified as having risks through the four-stream penetration correlation verification, assign a basic risk score to each identified risk behavior based on the risk type; if the number of consecutive occurrences of the risk behavior in historical data exceeds the preset threshold or the risk personnel exist in the historical risk list, then add an additional persistent risk score or recurring risk score.

[0110] The data risk identification intelligent agent acquires all identified risky behaviors and their corresponding risk types. For risky personnel identified through four-stream data penetration verification, the data risk identification intelligent agent assigns a basic risk score to each identified risky behavior based on the risk type and according to preset scoring rules. The preset scoring rules can be set according to the risk type, and the specific score values ​​can be flexibly adjusted according to actual circumstances; this application does not impose any limitations on this. For example, in some examples, the preset scoring rules are as follows: compliance risks (such as exceeding age limits, expired qualifications, etc.) 30-40 points per item; income anomaly risks (such as exceeding settlement limits, income deviating from the benchmark, etc.) 30-60 points per item; business authenticity risks (such as mismatch of four-stream data, data gaps, etc.) 40-60 points per item; and behavior frequency risks (such as high-frequency settlement, exceeding cross-regional limits, etc.) 30-40 points per item.

[0111] Then, the data risk identification agent retrieves historical risk personnel blacklists and historical risk control data of risk personnel. If a certain type of risk behavior occurs more than a preset threshold consecutively, such as abnormal income of the same type occurring for two consecutive months, 20 points of continuous risk score are added for each item. If the risk personnel are on the historical risk personnel blacklist, 30 points of recurrence risk score are added to strengthen the risk identification of repeated and continuous violations.

[0112] Step 2: Accumulate all risk scores of the risk personnel to obtain their total risk score.

[0113] The data risk identification intelligent agent automatically summarizes all basic and additional risk scores for each risky individual to obtain their total risk score. Simultaneously, it retains detailed scoring information, such as the basic risk score and the basis for additional risk scores for each risky behavior, ensuring that the scoring results are traceable and verifiable.

[0114] For example, if a risk personnel has two risk behaviors: failure to sign a settlement agreement (base score 40 points) and a mismatch between the settlement amount and the task acceptance results (base score 50 points), and the second risk behavior occurs continuously for 2 months (additional continuous risk score 20 points), then the risk personnel's total risk score is 40+50+20=110 points.

[0115] Step 3: Determine the risk level of the at-risk personnel based on the preset total score threshold range.

[0116] The data risk identification intelligent agent determines the risk level of each risk individual based on their total risk score and a preset total score threshold range. The total score threshold range can be flexibly adjusted according to actual circumstances, and this application does not impose any limitations on it. For example, in some examples, when a risk individual's total risk score is ≥80, the risk level is determined to be high risk; when the risk individual's total risk score is between 60 and 79, the risk level is determined to be medium risk; and when the risk individual's total risk score is <60, the risk level is determined to be low risk. After determining the risk level of each risk individual, the data risk identification intelligent agent can label each risk individual with the corresponding risk level, and simultaneously associate the scoring details and risky behaviors, providing a basis for subsequent risk report generation.

[0117] Thus, by constructing a quantitative risk scoring and grading system through a data risk identification intelligent agent, standardized risk scoring and risk level classification can be achieved for risk personnel confirmed through the four-stream verification process. This allows the platform to clearly understand the degree of violation and severity of harm caused by these personnel, providing data support for differentiated handling measures (such as suspension of settlement, deadline for re-certification, and personnel removal). Furthermore, by adding persistent and recurring risk scores, the identification and control of high-risk personnel can be strengthened, resolving the unreasonable problem of treating single violations and repeated violations equally in traditional screening, and accurately identifying core risk targets.

[0118] Figure 8 This is a schematic flowchart of step S1024 in the data risk identification method for flexible employment provided in an embodiment of this application. Figure 8 As shown, according to some embodiments of this application, optionally, S1024: generating a risk report containing information on at-risk personnel may include the following steps S801 to S804.

[0119] S801: Generate a structured list of risk personnel, which includes risk personnel information for at least one risk personnel. The risk personnel information includes at least one of the following: personnel identification, risk level, risk type, abnormal behavior, verification basis, and data links to related evidence.

[0120] The data risk identification intelligent agent integrates personnel information, multi-dimensional risk screening results, four-flow penetration correlation verification results, risk scores, and risk levels to generate a structured list of risky personnel according to a preset structured format. This list can include risk personnel information for at least one risky individual. When there are multiple risky individuals, their risk personnel information can be grouped by risk type and / or sorted by risk score or risk level.

[0121] Information on risk personnel can include at least one of the following: personnel identification, risk level, risk type, abnormal behavior, verification basis, and data links to related evidence. For example, personnel identification can include unique identification information such as personnel ID number or platform account, facilitating rapid identification of risk personnel. Risk types can include compliance-related, income anomaly-related, task volume-related, or behavior frequency-related. Abnormal behavior can include specific abnormal issues, such as non-contractual settlement or mismatch between settlement amount and task acceptance results. Verification basis can include corresponding verification steps and results, such as discrepancies between contract flow and business flow verification, or abnormal fund flow verification. Data links to related evidence can directly link to the corresponding original data from the four flows (flow, time, and logistics), acceptance vouchers, settlement details, and other related evidence, enabling risk traceability and verification.

[0122] S802: Generate a risk classification statistical report. The risk classification statistical report is used to perform quantity statistics and / or percentage analysis of identified risk personnel according to risk type and / or risk level.

[0123] The data risk identification intelligent agent generates a risk classification statistical report based on a list of risky personnel. This report can be used to statistically analyze the number and / or percentage of identified risky personnel according to risk type and / or risk level. In some embodiments, when statistically analyzing by risk type, the report may include the number, percentage, and distribution of key anomalies for each risk type (e.g., compliance, income, workload, behavior frequency). For example, in the compliance category, 60% of personnel may have failed qualifications. In other embodiments, when statistically analyzing by risk level, the report may include the number, percentage, and distribution of total risk scores for each risk level (e.g., high risk, medium risk, low risk). For example, in the high-risk category, 70% of personnel may have scores between 80 and 100, and 30% may have scores above 100. In still other embodiments, cross-statistics can be performed based on risk type and risk level, such as the percentage of personnel with abnormal behavior frequency in the high-risk category and the percentage of personnel with abnormal compliance in the medium-risk category.

[0124] Risk classification statistical reports can include visual results such as bar charts and pie charts to clearly present the overall risk situation, making it easier for the platform to quickly locate weak links in risk control and optimize risk control rules.

[0125] S803: Generate corresponding risk management suggestions for risk personnel with different risk levels or different risk types.

[0126] The data risk identification intelligence agent can generate corresponding risk handling suggestions based on different risk levels or risk types. For example, in some embodiments, for high-risk personnel, the generated risk handling suggestions could be to immediately suspend settlements, prioritize offline verification, and restrict order acceptance until verification is complete. For medium-risk personnel, the generated risk handling suggestions could be to provide supplementary materials within a specified period, conduct secondary verification, and track subsequent behavior. For low-risk personnel, the generated risk handling suggestions could be to remind them to rectify the situation and include them in routine monitoring.

[0127] For example, in some embodiments, based on risk type, for personnel with compliance-related anomalies, the generated risk handling suggestions could be updating qualifications or re-signing contracts. For personnel with revenue or workload-related anomalies, the generated risk handling suggestions could be reviewing settlement amounts or verifying the authenticity of tasks. For personnel with behavioral frequency-related anomalies, the generated risk handling suggestions could be investigating cases of proxy operations or fraudulent order acceptance. Furthermore, in addition to clearly defining operational steps, risk handling suggestions can also include the responsible department and completion deadline.

[0128] S804: Generate a risk report based on the list of at-risk personnel, risk classification statistical reports, and risk handling recommendations.

[0129] The data risk identification intelligent agent integrates the list of at-risk personnel, risk classification statistical reports, and risk handling suggestions, generating a risk report according to a preset format. The risk report may include the flexible employment project's identifier, the risk investigation timeframe, an investigation overview, the list of at-risk personnel, the risk classification statistical reports, and risk handling suggestions. The investigation overview may include the total number of people investigated, the total number of at-risk personnel, and the percentage of at-risk personnel. In some embodiments, the risk report supports online preview, export, and access-based viewing. The risk report can be synchronized to the flexible employment platform, triggering the risk handling assignment process, and supports linking to subsequent handling progress and result feedback, forming a closed-loop management system for report generation, handling execution, and result archiving. The risk report can include risk warnings, automatically pushing them to the corresponding responsible department heads for high-risk personnel, enabling rapid response and priority handling of high-risk situations.

[0130] Thus, by integrating the risk personnel list, risk classification statistical reports, and risk handling suggestions through a data risk identification intelligent agent to generate a risk report, the problem of fragmented risk information and lack of clear basis for risk control and review in flexible employment scenarios can be solved. The risk personnel list clearly presents the risk level, type, abnormal behavior, and verification basis for each risk personnel, allowing risk control personnel to quickly locate risk personnel and trace the root cause of risks. The risk classification statistical reports clearly show the distribution trends and core anomalies of different types and levels of risks, facilitating the platform's rapid identification of weak links in risk control. The risk handling suggestions provide accurate action guidance for risk control execution, facilitating the rapid handling of risk issues of different levels or types.

[0131] Optionally, according to some embodiments of this application, before S1022: performing multi-dimensional risk screening on personnel information and four-stream data based on a preset risk rule set, the data risk identification method for flexible employment further includes performing the following processing through a data risk identification intelligent agent: Based on the four-stream data of flexible employment projects within the risk screening time frame, calculate the risk assessment benchmark value for flexible employment projects; wherein, the risk assessment benchmark value includes at least one of the following: average income benchmark range, single task income benchmark value, and average task volume benchmark value.

[0132] Specifically, the data risk identification agent can retrieve four-stream data for flexible employment projects within the risk screening timeframe, and calculate a risk assessment benchmark value corresponding to the flexible employment project based on this four-stream data. This risk assessment benchmark value can be used as the judgment standard in the S1022 multi-dimensional risk screening. For example, the risk assessment benchmark value may include at least one of the following: average income benchmark range, single task income benchmark value, and average task volume benchmark value.

[0133] In this way, by using the four-stream data of the flexible employment project itself to calculate the risk assessment benchmark, the judgment of anomalies in the dimensions such as personnel income and workload can be more in line with the actual business characteristics of the project, reducing misjudgments caused by differences in business models, job types or salary standards between projects, such as normal behavior being misjudged as abnormal or abnormal behavior not being identified, and providing accurate and suitable judgment standards for multi-dimensional risk screening.

[0134] According to some embodiments of this application, optionally, S1024: generating a risk report containing information on at-risk personnel may further include the following steps: The calculated risk assessment benchmark value is compared with the preset industry benchmark data; If the deviation between the risk assessment benchmark and the corresponding industry benchmark data exceeds a preset deviation threshold, the risk assessment benchmark will be marked as abnormal in the risk report. The specific size of the deviation threshold can be flexibly adjusted according to the actual situation, and this application does not limit it.

[0135] Specifically, the data risk identification intelligent agent can compare the project risk assessment benchmark with preset industry benchmark data item by item, quantifying the degree of deviation between the two. For example, the industry benchmark data can be the average income benchmark range, single task income benchmark value, and / or average task volume benchmark value for flexible employment projects of the same category and scale. If the deviation of one or more risk assessment benchmark values ​​from the corresponding industry benchmark data exceeds a preset deviation threshold, the data risk identification intelligent agent can mark the abnormal risk assessment benchmark value as an anomaly in the final risk report generated, along with the specific deviation value, such as the single task income benchmark value of this project being 60% higher than the industry benchmark, exceeding the deviation threshold by 50%.

[0136] In this way, by marking the risk assessment benchmark values ​​that deviate from industry benchmark data in the risk report as abnormal, risk control personnel can know whether the project benchmark values ​​used in this risk screening are reasonable and whether they deviate from the industry's normal level. This allows them to promptly identify risk detection deviations that may be caused by abnormal benchmark value settings. At the same time, it provides clear guidance for risk control personnel to review the screening results and adjust and optimize the project risk assessment benchmark values, thereby improving the reliability of risk identification.

[0137] Based on the data risk identification method for flexible employment provided in any of the above embodiments, this application also provides a data risk identification system for flexible employment.

[0138] Figure 9 This is a structural block diagram of a data risk identification system for flexible employment provided in an embodiment of this application. Figure 9 As shown, the data risk identification system 90 for flexible employment provided in this application embodiment may include: Interaction module 901 is used to receive risk verification instructions input by the user. The risk verification instructions include the identifier of the flexible employment project and the time range for risk screening. Data risk identification agent 902 is used to perform the following processes: Based on the identification of flexible employment projects and the time range of risk screening, personnel information and four-flow data related to flexible employment projects are retrieved from multiple business systems. The four-flow data include contract flow data, business flow data, capital flow data and invoice flow data. Based on a pre-set set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, and they are marked as preliminary risk personnel. For personnel at initial risk, the personnel identifier of the personnel at initial risk and the task identifier of the task completed within the risk screening time range are used as the association key. Data records matching the association key are extracted from the four-stream data, and the extracted data records are subjected to four-stream penetration association verification to verify whether the personnel at initial risk are indeed at risk. Based on personnel information, multi-dimensional risk screening results, and four-flow penetration correlation verification results, a risk report containing information on at-risk personnel is generated.

[0139] The specific execution process of the interaction module 901 and the data risk identification agent 902 has been described in detail above and will not be repeated here.

[0140] The data risk identification system for flexible employment embodiments of this application, on the one hand, automatically retrieves personnel information and data from four flows—contract flow, business flow, capital flow, and invoice flow—based on the identification of flexible employment projects and the screening time range through a data risk identification intelligent agent. This eliminates the need for manual intervention, achieving automated linkage retrieval and integration of data from multiple systems. It solves the problems of data dispersion, low collaboration efficiency, and high labor costs in traditional models, ensuring the timeliness and completeness of data retrieval. On the other hand, based on the autonomous processing capabilities of the data risk identification intelligent agent, it autonomously identifies preliminary risk personnel based on compliance, income, workload, and behavioral frequency through multi-dimensional risk screening. Then, it automatically performs a four-flow data penetration-style correlation verification using personnel and task identifiers as association keys. This verifies the business authenticity, data consistency, and risk correlation of preliminary risk personnel, accurately identifying those who exceed age limits, have abnormally deviated income from the average, or have low workloads but high settlement amounts. Compared to manual screening and a single rule engine, this reduces the false positive and false negative rates, improving the accuracy and intelligence level of risk personnel identification. On the other hand, from data retrieval, multi-dimensional screening, in-depth verification to risk report generation, the entire process is completed autonomously by the data risk identification intelligent agent, realizing standardized closed-loop processing of the entire risk identification process. This not only adapts to the weekly and monthly routine risk control needs of flexible employment, compressing the traditional manual screening work that takes several days to complete in a short time, thus improving the efficiency of risk identification, but also reduces the reliance on human experience, while providing complete data support for subsequent risk review and rule optimization.

[0141] It should be noted that the data risk identification system 90 for flexible employment provided in this application embodiment has the same or corresponding technical features as the data risk identification method for flexible employment provided in any of the above embodiments, and produces the same technical effects. For the sake of brevity, it will not be described in detail here.

[0142] The electronic device in this application embodiment may be a user terminal device, a server, other computing devices, or a cloud server. Figure 10 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. The electronic device may include a processor 1001 and a memory 1002 storing computer program instructions. When the processor 1001 executes the computer program instructions, it implements the process or function of the data risk identification method for flexible employment in any of the above embodiments.

[0143] Specifically, processor 1001 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. Memory 1002 may include mass storage for data or instructions. For example, memory 1002 may be at least one of the following: hard disk drive (HDD), read-only memory (ROM), random access memory (RAM), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, universal serial bus (USB) drive, or other physical / tangible memory storage device. Alternatively, memory 1002 may include removable or non-removable (or fixed) media. Furthermore, memory 1002 may be internal or external to an electronic device. Memory 1002 may be non-volatile solid-state memory. In other words, typically memory 1002 includes a tangible (non-transitory) computer-readable storage medium (such as a memory device) encoded with computer-executable instructions, and when the software is executed (e.g., by one or more processors), it can perform the operations described in the methods of the embodiments of this application. The processor 1001 implements the process or function of any of the data risk identification methods for flexible employment in the above embodiments by reading and executing computer program instructions stored in the memory 1002.

[0144] In one example Figure 10The illustrated electronic device may also include a communication interface 1003 and a bus 1010. The processor 1001, memory 1002, and communication interface 1003 are connected via bus 1010 and communicate with each other. Communication interface 1003 is primarily used to enable communication between modules, devices, units, and / or equipment in the embodiments of this application. Bus 1010 includes hardware, software, or both, and can couple components of the online data traffic billing device together. For example, the bus may include at least one of the following: Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) Interconnect, Industry Standard Architecture (ISA) bus, Infinite Bandwidth Interconnect, Low Pin Count (LPC) bus, memory bus, Microchannel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus, or other suitable buses. Bus 1010 may include one or more buses. Although specific buses are described or illustrated in the embodiments of this application, any suitable bus or interconnection method may be considered in the embodiments of this application.

[0145] In conjunction with the methods in the above embodiments, this application also provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the process or function of any of the data risk identification methods for flexible employment in the above embodiments.

[0146] In addition, this application also provides a computer program product that stores computer program instructions. When the computer program instructions are executed by a processor, they implement the process or function of any of the data risk identification methods for flexible employment described above.

[0147] The flowcharts and / or block diagrams of methods, apparatuses, systems, and computer program products according to embodiments of this application have been exemplarily described above, and related aspects have been described. It should be understood that each block or combination thereof in the flowcharts and / or block diagrams may be implemented by computer program instructions, by dedicated hardware performing a specified function or action, or by a combination of dedicated hardware and computer instructions. For example, these computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to form a machine that enables the implementation of the function / action specified in each block or combination thereof in the flowcharts and / or block diagrams, executable via such processor. Such a processor may be a general-purpose processor, a dedicated processor, a special-purpose application processor, or a field-programmable logic circuit.

[0148] The functional blocks shown in the structural block diagrams of this application can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc.; when implemented in software, they are programs or code segments used to perform the required tasks. Programs or code segments can be stored in memory or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. Code segments can be downloaded via computer networks such as the Internet or intranets.

[0149] It should be noted that this application is not limited to the specific configurations and processes described above or shown in the figures. The above descriptions are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the described systems, devices, modules, or units can be referred to the corresponding processes in the method embodiments, and need not be repeated here. It should be understood that the scope of protection of this application is not limited thereto. Any person skilled in the art can conceive of various equivalent modifications or substitutions within the scope of the technology disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application.

Claims

1. A data risk identification method for flexible employment, characterized in that, include: Receive risk verification instructions input by the user, the risk verification instructions including the identifier of the flexible employment project and the risk investigation time range; The following processes are performed by the data risk identification agent: Based on the identification of flexible employment projects and the time range of risk screening, personnel information and four-flow data related to flexible employment projects are retrieved from multiple business systems. The four-flow data include contract flow data, business flow data, fund flow data and invoice flow data. Based on a pre-set set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, and they are marked as preliminary risk personnel. For personnel at initial risk, the personnel identifier of the personnel at initial risk and the task identifier of the task completed within the risk investigation time range are used as the association key. Data records matching the association key are extracted from the four-stream data, and the extracted data records are subjected to four-stream penetration association verification to verify whether the personnel at initial risk are at risk. Based on personnel information, multi-dimensional risk screening results, and four-flow penetration correlation verification results, a risk report containing information on at-risk personnel is generated.

2. The method according to claim 1, characterized in that, Perform four-stream penetration correlation verification on the extracted data records, including: Based on personnel and task identifiers, the contracting entity information and validity period information in the contract flow data are correlated and verified with the task recipient information and task dispatch time in the business flow data. The verification is performed to check whether the task recipient has a valid contract and whether the task scope is consistent, and to check whether the task dispatch time is within the contract validity period. Based on the task identifier, the task acceptance results and acceptance time in the business flow data are correlated and verified with the corresponding settlement amount, payment time and payee information in the fund flow data. The verification is performed to check whether the settlement amount matches the task acceptance results, whether the payment time is later than the task acceptance time, and whether the payee identity is consistent. Based on personnel and task identifiers, the signing records in the contract flow data, the task process records in the business flow data, the payment records in the fund flow data, and the invoice records in the invoice flow data are sequentially linked to construct a full-process data link from contract signing to invoice issuance, and to identify whether there are any missing or unconnectable data gaps in the full-process data link.

3. The method according to claim 2, characterized in that, Verify whether the personnel conducting the initial risk assessment are indeed at risk, including: When the task recipient does not have a valid contract and / or the task scope is inconsistent, or when the task assignment time is not within the contract validity period, the aforementioned preliminary risk personnel are determined to be risk personnel with contract compliance risks. When the settlement amount does not match the task acceptance results, or the payment time is earlier than the task acceptance time, or the identity of the payee is inconsistent, the aforementioned preliminary risk personnel are determined to be risk personnel with business authenticity risk or abnormal fund entity risk. When there are missing or disconnected data in the entire data chain, the personnel identified as having a risk to the integrity of the process are determined to be at risk.

4. The method according to claim 1, characterized in that, Based on a pre-defined set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify individuals exhibiting anomalies in compliance, income, workload, and / or behavioral frequency, including: Based on business flow data and cash flow data, calculate the unit task income of personnel; wherein, the unit task income is the ratio of settlement income to the amount of tasks completed; The unit task revenue is compared with the preset benchmark range of unit task revenue within the project; If a person's unit task income deviates from the benchmark range, or if a person's task volume is zero but their settlement income is not zero, then the person is determined to have an anomaly in the task volume dimension.

5. The method according to claim 1, characterized in that, Based on a pre-defined set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify individuals exhibiting anomalies in compliance, income, workload, and / or behavioral frequency, including: Based on the personnel information, verify whether the personnel's age is within the preset legal working age range, and / or verify whether the personnel's job qualifications are valid and within the certification period; Based on contract flow data and fund flow data, verify whether personnel have settlement records in an unsigned state or continuous settlement records after the contract expires; If any of the above verification conditions are met, the person is deemed to have an anomaly in the compliance dimension.

6. The method according to claim 1, characterized in that, Based on a pre-defined set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify individuals exhibiting anomalies in compliance, income, workload, and / or behavioral frequency, including: Based on the aforementioned cash flow data, calculate the personnel's income within the risk assessment timeframe; The deviation of the revenue is compared with the preset benchmark revenue value within the project. If the deviation between the income and the benchmark income value within the project exceeds a preset threshold, or if the income exceeds the preset upper limit of static settlement amount, then the person is determined to have an abnormality in the income dimension.

7. The method according to claim 1, characterized in that, Based on a pre-defined set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify individuals exhibiting anomalies in compliance, income, workload, and / or behavioral frequency, including: Based on business flow data and cash flow data, the statisticians determine the settlement frequency and / or service area switching frequency within a preset time window; Compare the settlement frequency and / or service area switching frequency with preset frequency thresholds; If a person's settlement frequency or service area switching frequency exceeds the corresponding frequency threshold, then the person is deemed to have an abnormality in the behavioral frequency dimension.

8. The method according to claim 1, characterized in that, Before generating the risk report, the method further includes: The following processes are performed by the data risk identification agent: For risk personnel identified as having risks through the four-stream penetration-based correlation verification, a basic risk score is assigned to each identified risk behavior based on the risk type; if the number of consecutive occurrences of a risk behavior in historical data exceeds a preset threshold or the risk personnel exist in the historical risk list, additional persistent risk scores or recurring risk scores are added. Accumulate all risk scores of the risk personnel to obtain their total risk score; The risk level of individuals is determined based on a preset total score threshold range.

9. The method according to claim 8, characterized in that, Generate a risk report containing information about at-risk individuals, including: Generate a structured list of risk personnel, which includes risk personnel information for at least one risk personnel. The risk personnel information includes at least one of the following: personnel identifier, risk level, risk type, abnormal behavior, verification basis, and data links to related evidence. Generate a risk classification statistical report, which is used to perform quantity statistics and / or percentage analysis on the identified risk personnel according to risk type and / or risk level; Generate corresponding risk management suggestions for risk personnel with different risk levels or different risk types; A risk report is generated based on the list of at-risk personnel, the risk classification statistical report, and the risk handling recommendations.

10. The method according to claim 1, characterized in that, Before performing multi-dimensional risk screening of personnel information and four-stream data based on a preset risk rule set, the method further includes: The following processes are performed by the data risk identification agent: Based on the four-stream data of the flexible employment project within the risk screening time range, the risk assessment benchmark value of the flexible employment project is calculated; wherein, the risk assessment benchmark value includes at least one of the following: average income benchmark range, single task income benchmark value, and average task volume benchmark value; Generating risk reports that include information about at-risk individuals also includes: The calculated risk assessment benchmark value is compared with the preset industry benchmark data; If the deviation between the risk assessment benchmark value and the corresponding industry benchmark data exceeds a preset deviation threshold, the risk assessment benchmark value will be marked as abnormal in the risk report.

11. A data risk identification system for flexible employment, characterized in that, include: The interaction module is used to receive risk verification instructions input by the user. The risk verification instructions include the identifier of the flexible employment project and the time range for risk screening. A data risk identification intelligent agent, which performs the following processes: Based on the identification of flexible employment projects and the time range of risk screening, personnel information and four-flow data related to flexible employment projects are retrieved from multiple business systems. The four-flow data include contract flow data, business flow data, fund flow data and invoice flow data. Based on a pre-set set of risk rules, multi-dimensional risk screening is performed on personnel information and four-stream data to identify personnel with abnormalities in the dimensions of compliance, income, workload and / or behavior frequency, and they are marked as preliminary risk personnel. For personnel at initial risk, the personnel identifier of the personnel at initial risk and the task identifier of the task completed within the risk investigation time range are used as the association key. Data records matching the association key are extracted from the four-stream data, and the extracted data records are subjected to four-stream penetration association verification to verify whether the personnel at initial risk are at risk. Based on personnel information, multi-dimensional risk screening results, and four-flow penetration correlation verification results, a risk report containing information on at-risk personnel is generated.

12. An electronic device, characterized in that, The electronic device includes a processor and a memory storing computer program instructions; when the electronic device executes the computer program instructions, it implements the data risk identification method for flexible employment as described in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the data risk identification method for flexible employment as described in any one of claims 1-10.

14. A computer program product, characterized in that, It includes computer program instructions that, when executed by a processor, implement the data risk identification method for flexible employment as described in any one of claims 1-10.

Images