Campus resource processing system for idle article lease and group purchase

Abstract

The invention discloses a campus resource processing system for idle article leasing and group buying, and the system is compared with static authentication, evaluation scoring and manual processing strategies which are commonly adopted by an existing campus resource leasing or group buying platform. According to the campus resource processing system provided by the invention, the integrated authentication module, the behavior analysis module, the credit scoring module, the risk early warning module, the grade adjustment module and the feedback module are integrated to construct a complete data-driven closed-loop system; according to the system, through multi-dimensional mechanisms such as identity credibility guarantee, behavior dynamic perception, credit quantitative evaluation and risk linkage control, the technical defects of an existing platform in the aspects of trust construction, user governance and transaction risk control are effectively overcome, and the safety, the intelligent level and the operation efficiency of the system are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of smart campus information management technology, and in particular to a campus resource processing system for the rental and group purchase of idle items. Background Technology

[0002] Currently, in university settings, with the popularization of the sharing economy concept and the advancement of campus digitalization, the demand for renting idle items such as textbooks, school supplies, and household appliances among students, as well as group buying based on social trust, is growing.

[0003] However, existing campus resource trading platforms generally suffer from the following technical and management deficiencies: First, the lack of user identity verification mechanisms makes it difficult for platforms to distinguish between users inside and outside the campus, posing risks to information security and the authenticity of trading entities; second, the lack of effective collection and analysis mechanisms for user behavior data makes it difficult for platforms to accurately identify user activity and participation, hindering the implementation of tiered management; third, the lack of scientific credit evaluation models and risk control measures leads to frequent occurrences of transaction defaults, malicious reviews, and frequent reports, seriously affecting the platform's credibility and user stickiness; and fourth, the vague user level system and lagging feedback mechanisms prevent the realization of dynamic management and risk warning based on behavior and credit. Summary of the Invention

[0004] The main purpose of this application is to provide a campus resource processing system for the rental and group purchase of idle items, so as to solve the problems mentioned in the background.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] A campus resource processing system for the rental and group purchase of idle goods includes an authentication module, a behavior analysis module, a credit scoring module, a risk warning module, a level adjustment module, and a feedback module.

[0007] The authentication module is used to collect and authenticate user information;

[0008] The behavior analysis module is used to analyze user behavior and determine activity levels;

[0009] The credit scoring module is used to analyze user credit and confirm credit behavior;

[0010] The risk warning module is used to analyze user operations and identify operational risks;

[0011] The rating adjustment module is used to adjust user ratings and optimize user classification;

[0012] The feedback module is used to provide feedback on the collected parameters and analysis results.

[0013] Preferably, the authentication module includes an authentication data acquisition unit and an authentication analysis unit;

[0014] The authentication data acquisition unit is used to collect multi-source data for user authentication, and to preprocess and dimensionless the collected multi-source data, including the number of authentication factors (A), data completeness (B), number of information matching items (C), IP validity (D), and device change rate (E).

[0015] The authentication analysis unit is used to couple the collected multi-source data, input the processed multi-data into a pre-trained deep learning framework, perform feature fusion through a multi-layer neural network, calculate and obtain the authentication analysis coefficient RZF, perform data analysis on the calculated authentication analysis coefficient RZF, and determine whether the user meets the authentication requirements based on the analysis results.

[0016] The authentication analysis unit calculates the authentication analysis coefficient RZF using the following formula;

[0017] ;

[0018] In the formula: A is the number of certification factors, B is the data completeness, C is the number of information matching items, D is the IP validity, and E is the equipment change rate.

[0019] Preferably, the authentication analysis unit performs the authentication analysis coefficient RZF analysis in the following specific way:

[0020] when When this message appears, it means the current user does not meet the usage requirements and cannot register.

[0021] when When this time is reached, it means that the current user meets the usage requirements and can register.

[0022] Preferably, the behavior analysis module includes a behavior data acquisition unit and a behavior analysis unit;

[0023] The behavior data collection unit is used to collect user behavior data and preprocess and dimensionless the collected data, including daily login frequency F, number of effective operations G, average dialogue market H, active day ratio I, and function coverage J.

[0024] The behavior analysis unit is used to couple the collected data, input the processed multiple data into a pre-trained deep learning framework, perform feature fusion through a multi-layer neural network, obtain behavior analysis coefficients, perform data analysis on the behavior analysis coefficients, and determine user activity based on the analysis results.

[0025] The behavior analysis unit calculates and obtains the behavior analysis coefficients using the following formula;

[0026]

[0027] In the formula: F is the daily login frequency, G is the number of effective operations, H is the average conversation duration, I is the percentage of active days, and J is the feature coverage rate.

[0028] Preferably, the behavior analysis unit performs the behavior analysis coefficients in the following specific manner:

[0029] when This indicates that the current user is a Level 3 activity user, which is considered a low-activity user.

[0030] when At this time, it indicates that the current user is a level 2 active user, which is considered a medium-active user;

[0031] when At this time, it means that the current user is a level 1 active user, which is considered a highly active user.

[0032] Preferably, the credit scoring module includes a credit data collection unit and a credit analysis unit;

[0033] The credit data collection unit is used to collect users' credit data and preprocess and dimensionless the collected data, including default rate K, user rating value L, number of reports M, number of completed transactions N, and positive review rate O.

[0034] The credit analysis unit is used to couple the processed data, input the processed multiple data into a pre-trained deep learning framework, perform feature fusion through a multi-layer neural network, calculate and obtain the credit analysis coefficient, perform data analysis on the calculated credit analysis coefficient, and determine the user's credit rating based on the analysis results.

[0035] The credit analysis unit calculates the credit analysis coefficient using the following formula;

[0036]

[0037] In the formula: K is the default rate, L is the user rating, M is the number of reports, N is the number of completed transactions, and O is the positive review rate.

[0038] Preferably, the specific analysis method of the credit analysis unit for the credit analysis coefficient is as follows:

[0039] when At this time, it indicates that the current user is a Level 3 credit user, which is a low credit rating;

[0040] when At this time, it indicates that the current user is a level 2 credit user, with a medium credit rating;

[0041] when At this time, it indicates that the current user is a Level 1 credit user, with a high level of credit.

[0042] Preferably, the risk warning module integrates and analyzes the analysis results of the behavior analysis module and the credit scoring module to determine whether there are any abnormalities in the user's transactions. The specific analysis method is as follows;

[0043] If the current user's behavior analysis result is Level 1 and the credit analysis result is Level 1, it means that the current user's transaction credibility is Level 1 and no monitoring or prompts are required.

[0044] If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0045] If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0046] If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level two and needs to be monitored.

[0047] If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level three, which requires monitoring and notification to the other party in the transaction.

[0048] If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required.

[0049] If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0050] If the current user's behavior analysis result is level three and the credit analysis result is level two, it means that the current user's transaction credibility is level two and needs to be monitored.

[0051] If the current user's behavior analysis result is level three, and the credit analysis result is also level three, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required.

[0052] Preferably, the level adjustment module integrates and analyzes the analysis results from the behavior analysis module, the credit scoring module, and the risk warning module, and determines the user level based on the analysis results. The specific method is as follows:

[0053] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 1, it means that the current user is a Level 1 user and belongs to a high-quality user.

[0054] If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level one user and belongs to a high-quality user.

[0055] If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level two user, belonging to the medium quality user category;

[0056] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 2, it means that the current user is a Level 1 user and belongs to a high-quality user.

[0057] If the current user's behavior analysis result is Level 2, and the transaction credibility is Level 2, it means that the current user is a Level 2 user, belonging to the medium-quality user category;

[0058] If the current user's behavior analysis result is level three and the transaction credibility is level two, it means that the current user is a level three user, belonging to a low-quality user.

[0059] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 3, it means that the current user is a Level 3 user, which is a low-quality user.

[0060] If the current user's behavior analysis result is level two and the transaction credibility is level three, it means that the current user is a special user and needs to be specially marked.

[0061] If the current user's behavior analysis result is level three, and the transaction credibility is also level three, it means that the current user is a special user and needs to be specially marked. Attached Figure Description

[0062] Figure 1 This is the system flowchart for this application.

[0063] In the diagram: 1. Authentication module; 11. Authentication data collection unit; 12. Authentication analysis unit; 2. Behavior analysis module; 21. Behavior data collection unit; 22. Behavior analysis unit; 3. Credit scoring module; 31. Credit data collection unit; 32. Credit analysis unit; 4. Risk warning module; 5. Rating adjustment module; 6. Feedback module. Detailed Implementation

[0064] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0065] The terms "first," "second," and "third" in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0066] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0067] Example 1: Please refer to Figure 1 A campus resource processing system for the rental and group purchase of idle goods includes an authentication module 1, a behavior analysis module 2, a credit scoring module 3, a risk warning module 4, a level adjustment module 5, and a feedback module 6.

[0068] Authentication module 1 is used to collect and authenticate user information;

[0069] Behavior analysis module 2 is used to analyze user behavior and determine activity levels;

[0070] Credit scoring module 3 is used to analyze user credit and confirm credit behavior;

[0071] Risk warning module 4 is used to analyze user operations and identify operational risks;

[0072] The rating adjustment module 5 is used to adjust user ratings and optimize user classification;

[0073] Feedback module 6 is used to provide feedback on the collected parameters and analysis results.

[0074] In this embodiment, authentication module 1 is used to collect, verify, and determine user identity information. This module integrates with various authentication methods such as the campus card system, academic affairs system, campus email, and facial recognition, combined with IP validity detection and device stability analysis, to achieve multi-source information fusion and a multi-factor authentication mechanism. Authentication module 1 can effectively identify and restrict access permissions for off-campus users, ensuring that platform users are genuine, unique, and in a valid student status. This provides a foundation for identity protection for subsequent behavior assessment, credit evaluation, and risk control, effectively enhancing the platform's trust boundary and security level.

[0075] The behavior analysis module 2 is used to collect key behavioral data of users during their use of the platform and to quantitatively analyze their activity levels. The behavior data collection unit 21 collects behavioral characteristics such as average daily login frequency, number of effective operations, average session duration, percentage of active days, and feature usage coverage. The behavior analysis unit 22 then performs dimensionless and coupled processing on the data to form behavior analysis coefficients for activity level classification. The behavior analysis module 2 can not only identify high-frequency active users and potentially inactive users, but also provide data support for subsequent credit scoring and platform operation strategies, enhancing the platform's user behavior insight capabilities and service accuracy.

[0076] Credit scoring module 3 is used to analyze and evaluate users' credit behavior during their use of the platform. Credit data collection unit 31 collects credit-related parameters such as user default rate, user rating, number of reports received, number of completed transactions, and positive review rate. Credit analysis unit 32 then calculates credit analysis coefficients and classifies users' credit levels accordingly. Credit scoring module 3 enables dynamic management of users' credit status, helping to enhance the platform's incentive effect on trustworthy users and the punishment for dishonest behavior, thereby establishing a robust credit system based on behavioral patterns and historical records, and optimizing the platform's transaction order and trust structure.

[0077] The risk warning module 4 is used to identify and issue warnings for the risk level of users' transaction operations based on behavioral analysis results and credit scoring results. This module integrates the results from the behavioral analysis module 2 and the credit scoring module 3 through rule logic or algorithm models to dynamically assess the user's risk level and trigger prompts, operation restrictions, or third-party intervention mechanisms accordingly. The risk warning module 4 can automatically respond to situations such as frequent device switching, abnormal transaction behavior, or significant credit fluctuations, effectively preventing malicious operations, default transactions, or fraudulent activities, and improving the platform's risk control automation level and processing timeliness.

[0078] The Level Adjustment Module 5 dynamically adjusts user classification levels based on user behavior analysis, credit scores, and risk level results. This module comprehensively judges user status according to preset level adjustment logic, classifying users into Level 1, Level 2, Level 3, or specially marked users, and links it with platform control items such as usage permissions, credit policies, and resource recommendations. Level Adjustment Module 5 enhances the platform's ability to manage user differentiation, realizing a shift from static role allocation to dynamic behavior-driven management, and strengthening the platform's operational adaptability and flexibility.

[0079] Feedback module 6 is used to summarize and transmit various collected parameters and analysis results, and to enable result feedback between the user end and the platform management end. This module can provide feedback on key indicators or evaluation results obtained from authentication module 1, behavior analysis module 2, credit scoring module 3, risk warning module 4, and level adjustment module 5 through visualization or interface push. Feedback module 6 not only improves the transparency and user awareness of system operation, but also provides support for platform strategy optimization, data closed-loop adjustment, and continuous operation, constructing a processing structure that combines system intelligence and user-friendly interaction.

[0080] Compared to the static authentication, evaluation scoring, and manual processing strategies commonly used by existing campus resource rental or group-buying platforms, the campus resource processing system proposed in this invention constructs a complete data-driven closed-loop system by integrating six functional modules: authentication module 1, behavior analysis module 2, credit scoring module 3, risk warning module 4, level adjustment module 5, and feedback module 6. Through multi-dimensional mechanisms such as identity credibility assurance, dynamic behavior perception, quantitative credit assessment, and risk linkage control, this system effectively compensates for the technical shortcomings of existing platforms in trust building, user governance, and transaction risk control, significantly improving the system's security, intelligence level, and operational efficiency.

[0081] Example 2: Please refer to Figure 1 The authentication module 1 includes an authentication data acquisition unit 11 and an authentication analysis unit 12;

[0082] The authentication data acquisition unit 11 is used to collect multi-source data for user authentication, and to preprocess and dimensionless the collected multi-source data, including the number of authentication factors (A), data completeness (B), number of information matching items (C), IP validity (D), and equipment change rate (E).

[0083] The authentication analysis unit 12 is used to couple the collected multi-source data, calculate and obtain the authentication analysis coefficient RZF, perform data analysis on the calculated authentication analysis coefficient RZF, and determine whether the user meets the authentication requirements based on the analysis results.

[0084] The authentication analysis unit 12 calculates the authentication analysis coefficient RZF using the following formula;

[0085] ;

[0086] In the formula: A is the number of certification factors, B is the data completeness, C is the number of information matching items, D is the IP validity, and E is the equipment change rate.

[0087] The specific analysis method of the certification analysis coefficient RZF by certification analysis unit 12 is as follows:

[0088] when When this message appears, it means the current user does not meet the usage requirements and cannot register.

[0089] when When this time is reached, it means that the current user meets the usage requirements and can register.

[0090] In this embodiment, the authentication data acquisition unit 11 not only collects traditional identity fields such as data completeness (B) and information matching items (C), but also introduces dynamic and behavioral authentication elements such as the number of authentication factors (A), IP validity (D), and device change rate (E). Through the normalization and preprocessing of the aforementioned multi-source data, the accuracy, timeliness, and risk identification capabilities of the authentication information are improved, solving the problems of single authentication methods and weak anti-impersonation capabilities in existing technologies.

[0091] The authentication analysis unit 12 employs a structured mathematical model to weightedly couple multiple indicators, calculating the authentication analysis coefficient RZF to quantify the user's identity credibility. This analysis formula considers multiple dimensions, including identity strength, behavioral stability, and device consistency, and the output results have clear numerical threshold criteria. By setting authentication thresholds, the system can automatically determine registered users, thereby reducing manual review costs and enhancing the objectivity and transparency of the authentication logic.

[0092] By setting numerical judgment rules for the authentication analysis coefficient RZF, when RZF ≤ 0.65, the current user is deemed not to meet the platform's usage requirements and their registration is rejected; when RZF > 0.65, the user's authentication is deemed qualified and registration is allowed. This mechanism transforms the previous identity verification process, which relied on subjective judgment, into a data-driven, automated decision-making access model, greatly improving the objectivity and consistency of the authentication process.

[0093] RZF's automated decision-making mechanism enables efficient screening of large numbers of users, completing initial access verification without manual intervention. This approach is suitable for university platforms facing scenarios involving large-scale new student registrations or periodic user identity changes, effectively reducing management burden and improving system response speed and user access experience.

[0094] By comprehensively calculating RZF values ​​and setting thresholds, users with incomplete identity information, abnormal behavioral characteristics, or potential device usage risks can be accurately identified, preventing unauthorized personnel, fake accounts, or malicious users from entering the platform. This mechanism helps build a highly trustworthy and clean campus user ecosystem, laying a secure foundation for subsequent behavior assessment and credit mechanisms.

[0095] The introduction of this module enables the platform to effectively filter out fake, temporary, or ambiguous registered users, enhancing the closed nature of the system boundaries and the purity of internal resource sharing. It is particularly suitable for campus scenarios where registration and use are restricted to on-campus personnel, ensuring the legitimacy of platform members' identities and providing a reliable data foundation for subsequent behavior assessment, credit modeling, and risk control mechanisms.

[0096] Example 3: Please refer to Figure 1 The behavior analysis module 2 includes a behavior data acquisition unit 21 and a behavior analysis unit 22;

[0097] The behavior data collection unit 21 is used to collect user behavior data and preprocess and dimensionless the collected data, including daily login frequency F, number of effective operations G, average dialogue market H, active day ratio I, and function coverage J.

[0098] The behavior analysis unit 22 is used to couple the collected data to obtain behavior analysis coefficients, perform data analysis on the behavior analysis coefficients, and determine user activity based on the analysis results.

[0099] Behavior analysis unit 22 calculates and obtains behavior analysis coefficients using the following formula;

[0100] ;

[0101] In the formula: F is the daily login frequency, G is the number of effective operations, H is the average conversation duration, I is the percentage of active days, and J is the feature coverage rate.

[0102] The behavior analysis unit 22 specifically analyzes the behavior analysis coefficient as follows:

[0103] When it is the case, it represents that the current user is a user with a third-level activity level and belongs to a low-active user;

[0104] When it is the case, it represents that the current user is a user with a second-level activity level and belongs to a medium-active user;

[0105] When it is the case, it represents that the current user is a user with a first-level activity level and belongs to a highly active user.

[0106] In this embodiment: The behavior data collection unit 21 not only collects conventional behavior indicators such as the daily login frequency F and the number of effective operations G of the user, but also introduces in-depth usage features such as the average session duration H, the proportion of active days I, and the function coverage rate J, thus forming a relatively comprehensive user behavior portrait system. By preprocessing and dimensionlessizing these parameters, the equal-scale fusion of behavior data in different dimensions is achieved, providing a unified input basis for subsequent calculations and overcoming the evaluation deviation problem caused by single index dimension or inconsistent standards in existing behavior analysis.

[0107] The behavior analysis unit 22 generates a behavior analysis coefficient XWF by constructing a weighted calculation model and fusing the above standardized behavior parameters. This model can perform systematic coupling calculations based on the user's login frequency, operation activity level, usage depth, and function breadth on the platform, thereby achieving a scientific classification of user activity levels. Compared with traditional static or empirical judgments, the present invention realizes a calculable, interpretable, and traceable behavior activity evaluation system.

[0108] By setting a numerical interval determination criterion, the behavior analysis unit 22 can classify users into three categories according to the calculation result of XWF: when XWF ≤ 0.75, the user is classified as a third-level activity user; when 0.75 < XWF ≤ 1, it is a second-level activity user; when XWF > 1, it is a first-level activity user. This classification criterion has clear numerical boundaries, overcomes the problems of fuzzy boundaries and subjective determination in traditional activity recognition, and improves the accuracy and stability of user status recognition.

[0109] The active performances of users at different levels on the platform are significantly different. The classification results of the activity levels can be used to drive the differential execution of subsequent operation strategies such as resource allocation, recommendation ranking, and notification push. For example, new resources can be preferentially pushed to highly active users and higher interaction permissions can be granted; content awakening or function guiding measures can be implemented for medium and low active users, thereby improving the overall active level and user retention rate of the platform.

[0110] By using a quantifiable grading system, the platform can dynamically identify and accurately respond to user lifecycle stages, providing a basis for user governance based on behavioral dimensions and addressing the shortcomings of traditional systems that lack behavioral status determination mechanisms.

[0111] Based on the activity level results output by Behavior Analysis Module 2, the platform can further implement differentiated operational strategies, such as dynamically adjusting user permissions, pushing personalized content, and implementing tiered incentive measures, to promote long-term user retention and increase platform stickiness. Simultaneously, this module also provides behavioral input for subsequent credit scoring and risk control, achieving a data loop from "credible identity" to "perceptible behavior."

[0112] Example 4: Please refer to Figure 1 The credit scoring module 3 includes a credit data collection unit 31 and a credit analysis unit 32;

[0113] The credit data collection unit 31 is used to collect users' credit data and preprocess and dimensionless the collected data, including default rate K, user rating value L, number of reports M, number of transactions completed N, and positive review rate O.

[0114] The credit analysis unit 32 is used to couple the processed data, calculate and obtain the credit analysis coefficient, perform data analysis on the calculated credit analysis coefficient, and determine the user's credit rating based on the analysis results.

[0115] Credit analysis unit 32 calculates the credit analysis coefficient using the following formula;

[0116] ;

[0117] In the formula: K is the default rate, L is the user rating, M is the number of reports, N is the number of completed transactions, and O is the positive review rate.

[0118] The specific analysis method for the credit analysis coefficient in credit analysis unit 32 is as follows:

[0119] when At this time, it indicates that the current user is a Level 3 credit user, which is a low credit rating;

[0120] when At this time, it indicates that the current user is a level 2 credit user, with a medium credit rating;

[0121] when At this time, it indicates that the current user is a Level 1 credit user, with a high level of credit.

[0122] In this embodiment: The credit data collection unit 31 not only collects positive credit indicators such as the default rate K, evaluation average L, and positive review rate O of users during the transaction process, but also incorporates negative or supplementary indicators such as the number of reports M and the number of completed transactions N, forming a comprehensive and rich-dimensional credit behavior data set. Through unified standardization processing, dimensionless and comparable data of various types are achieved, laying a stable foundation for subsequent model calculations.

[0123] The credit analysis unit 32 uses a structured formula to perform weighted coupling on various credit parameters, generating a credit analysis coefficient XYF, which is then used for the automatic classification of user credit levels. This model, through technical means such as reverse index processing of the default rate and exponential function suppression of the impact of reports, takes into account the stability of historical behavior and sudden risk factors, effectively avoiding the unbalanced impact of extreme behaviors on the score, and achieving a more scientific and reasonable credit scoring result.

[0124] According to the specific numerical range of the credit analysis coefficient XYF, the present invention divides users into three levels of credit ratings: when XYF ≤ 0.55, it is a third-level credit user; when 0.55 < XYF ≤ 0.75, it is a second-level credit user; when XYF > 0.75, it is a first-level credit user. This classification method is based on the model calculation results, with clear numerical boundaries and clear classification logic, effectively solving the problems of strong subjectivity and non-traceability in the credit rating setting of the existing system.

[0125] Through the accurate classification of user credit levels, the platform can incorporate users with different credit levels into differential management strategies. For example, first-level credit users can enjoy positive incentives such as deposit reduction and priority exposure; while second-level or third-level credit users can be subject to risk warnings, restricted trading permissions or set credit observation periods, so as to achieve the unity of credit guidance, risk containment and platform order maintenance.

[0126] This level determination mechanism is based on data-driven and model output, enabling the platform to have the ability of "measurable, reproducible, and intervenable" in credit management. Users can also adjust their platform behaviors according to their own credit levels and changing trends, thus forming a virtuous interactive credit ecological closed loop of user self-governance and platform governance.

[0127] Based on the results of XYF, the platform can divide users into high-credit, medium-credit and low-credit levels, and implement differential management measures accordingly, such as deposit exemption, limit restriction, trading recommendation ranking, etc. At the same time, the system can continuously update the credit score to dynamically identify potential risk users, providing input support for the risk warning module, and realizing the linkage closed loop from "behavior → credit → risk control".

[0128] Embodiment Five: Please refer to Figure 1The risk warning module 4 integrates and analyzes the results of the behavior analysis module 2 and the credit scoring module 3 to determine whether there are any abnormalities in the user's transactions. The specific analysis method is as follows:

[0129] If the current user's behavior analysis result is Level 1 and the credit analysis result is Level 1, it means that the current user's transaction credibility is Level 1 and no monitoring or prompts are required.

[0130] If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0131] If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0132] If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level two and needs to be monitored.

[0133] If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level three, which requires monitoring and notification to the other party in the transaction.

[0134] If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required.

[0135] If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required.

[0136] If the current user's behavior analysis result is level three and the credit analysis result is level two, it means that the current user's transaction credibility is level two and needs to be monitored.

[0137] If the current user's behavior analysis result is level three, and the credit analysis result is also level three, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required.

[0138] In this embodiment, a dynamic assessment and tiered response mechanism for user transaction credibility is constructed by designing a joint judgment logic model based on the analysis results of the behavior analysis module 2 and the credit scoring module 3 in the risk warning module 4. This mechanism performs a matrix-style combination judgment based on the user's active behavior level and credit score level on the platform to generate a transaction credibility level, and implements different levels of risk monitoring or transaction intervention accordingly.

[0139] By jointly analyzing behavioral activity and credit performance, a more comprehensive user risk profile is created. This dual-source fusion mechanism effectively avoids misjudgments or omissions caused by relying solely on credit or behavior, enhancing the coverage and accuracy of risk identification.

[0140] The risk warning module 4 categorizes transaction credibility into four levels based on the combination of user behavior analysis results and credit score results, and specifies corresponding handling methods, including multi-level response measures such as "no monitoring required," "recommended monitoring," "strong notification to the other party," and "restricting transactions and introducing third-party authentication." This structured processing flow effectively improves the automation level of risk control, avoiding the problems of delayed manual judgment and inconsistent responses. By setting clear credibility levels and corresponding strategies, when a user's activity level and credit level are on the edge, the system can automatically identify it as a high-risk combination, triggering timely warnings or transaction restrictions to prevent malicious operations or transactions that overdraw trust, thereby significantly enhancing the platform's security resilience.

[0141] This invention constructs a transaction credibility identification mechanism that is hierarchical, responsive, and comprehensive by integrating the analysis results of behavior and credit in the risk warning module 4. It overcomes the limitations of existing systems, such as lagging risk control, single rules, and lack of dynamic adaptation. It significantly improves the platform's ability to identify and intervene in risky users and transaction behaviors, and provides a strong risk control guarantee for building a credible, safe, and efficient campus resource transaction environment.

[0142] Example 6: Please refer to Figure 1 The rating adjustment module 5 integrates and analyzes the results from the behavior analysis module 2, credit scoring module 3, and risk warning module 4. Based on the analysis results, it determines the user rating, as follows:

[0143] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 1, it means that the current user is a Level 1 user and belongs to a high-quality user.

[0144] If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level one user and belongs to a high-quality user.

[0145] If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level two user, belonging to the medium quality user category;

[0146] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 2, it means that the current user is a Level 1 user and belongs to a high-quality user.

[0147] If the current user's behavior analysis result is Level 2, and the transaction credibility is Level 2, it means that the current user is a Level 2 user, belonging to the medium-quality user category;

[0148] If the current user's behavior analysis result is level three and the transaction credibility is level two, it means that the current user is a level three user, belonging to a low-quality user.

[0149] If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 3, it means that the current user is a Level 3 user, which is a low-quality user.

[0150] If the current user's behavior analysis result is level two and the transaction credibility is level three, it means that the current user is a special user and needs to be specially marked.

[0151] If the current user's behavior analysis result is level three, and the transaction credibility is also level three, it means that the current user is a special user and needs to be specially marked.

[0152] In this embodiment: the present invention constructs an automated classification and dynamic adjustment system for user levels by setting up a level adjustment module 5 and introducing a three-dimensional comprehensive evaluation mechanism based on the results of the behavior analysis module 2, credit scoring module 3 and risk warning module 4.

[0153] The rating adjustment module 5 integrates multiple indicators such as user activity, credit performance, and transaction risk level to establish a unified judgment logic rule. This rule not only considers user usage frequency and transaction volume but also incorporates an assessment of user behavior quality and historical credit, thereby constructing a comprehensive, reasonably weighted, and logically clear user rating system that significantly improves the accuracy and scientific nature of user classification.

[0154] This invention categorizes users into four levels: Level 1, Level 2, Level 3, and Special Users, and implements differentiated management based on platform strategies. For example, higher-level users enjoy greater access privileges, lower deposit thresholds, or priority referrals; lower-level or special users may have some functions restricted, receive guidance measures, or enter a observation period. This mechanism provides the platform with the technical foundation for matching resources and services according to user levels, enhancing the efficiency of platform resource allocation and operational precision.

[0155] This grading mechanism is dynamically adjustable; user levels are automatically updated based on behavioral changes, credit fluctuations, and risk control results, truly achieving real-time evolution of user profiles and dynamic adaptation to platform strategies. Compared to static classification methods, this approach better reflects users' current status and trend changes, helping to promptly identify potential risk groups and key targets for development.

[0156] This invention establishes a user hierarchical governance mechanism based on multidimensional data, with logical judgment as the core and response strategy as the guide through the level adjustment module 5. It breaks through the technical bottlenecks of existing platforms, such as extensive user management, distorted level division, and rigid service response, and significantly improves the platform's capabilities in terms of accurate user profiling, intelligent resource distribution, and dynamic operation management.

[0157] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0158] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A campus resource processing system for the rental and group purchase of idle goods, characterized by: It includes an authentication module (1), a behavior analysis module (2), a credit scoring module (3), a risk warning module (4), a rating adjustment module (5), and a feedback module (6). The authentication module (1) is used to collect and authenticate user information; The behavior analysis module (2) is used to analyze user behavior and confirm activity levels; The credit scoring module (3) is used to analyze user credit and confirm credit behavior; The risk warning module (4) is used to analyze user operations and identify operational risks; The rating adjustment module (5) is used to adjust user ratings and optimize user classification; The feedback module (6) is used to provide feedback on the collected parameters and analysis results.

2. The campus resource processing system for leasing and group buying of idle goods according to claim 1, characterized in that, The authentication module (1) includes an authentication data acquisition unit (11) and an authentication analysis unit (12); The authentication data acquisition unit (11) is used to collect multi-source data for user authentication and to preprocess and dimensionless the collected multi-source data, including the number of authentication factors A, data completeness B, number of information matching items C, IP validity D, and equipment change rate E. The authentication analysis unit (12) is used to couple the collected multi-source data, input the processed multi-data into a pre-trained deep learning framework, perform feature fusion through a multi-layer neural network, calculate and obtain the authentication analysis coefficient RZF, perform data analysis on the calculated authentication analysis coefficient RZF, and determine whether the user meets the authentication requirements based on the analysis results. The authentication analysis unit (12) calculates the authentication analysis coefficient RZF using the following formula; ； In the formula: A is the number of certification factors, B is the data completeness, C is the number of information matching items, D is the IP validity, and E is the equipment change rate.

3. The campus resource processing system for leasing and group buying of idle goods according to claim 2, characterized in that, The specific analysis method of the certification analysis coefficient RZF by the certification analysis unit (12) is as follows: when When this message appears, it means the current user does not meet the usage requirements and cannot register. when When this time is reached, it means that the current user meets the usage requirements and can register.

4. The campus resource processing system for leasing and group buying of idle goods according to claim 3, characterized in that, The behavior analysis module (2) includes a behavior data acquisition unit (21) and a behavior analysis unit (22). The behavior data collection unit (21) is used to collect user behavior data and preprocess and dimensionless the collected data, including daily login frequency F, number of effective operations G, average dialogue market H, active day ratio I and function coverage J. The behavior analysis unit (22) is used to couple the collected data, input the processed multiple data into the pre-trained deep learning framework, perform feature fusion through multi-layer neural networks, obtain behavior analysis coefficients, perform data analysis on the behavior analysis coefficients, and determine user activity based on the analysis results. The behavior analysis unit (22) calculates and obtains the behavior analysis coefficients using the following formula; ； In the formula: F is the daily login frequency, G is the number of effective operations, H is the average conversation duration, I is the percentage of active days, and J is the feature coverage rate.

5. The campus resource processing system for leasing and group buying of idle goods according to claim 4, characterized in that, The behavior analysis unit (22) analyzes the behavior analysis coefficients in the following specific way: when This indicates that the current user is a Level 3 activity user, which is considered a low-activity user. when At this time, it indicates that the current user is a level 2 active user, which is considered a medium-active user; when At this time, it means that the current user is a level 1 active user, which is considered a highly active user.

6. The campus resource processing system for leasing and group buying of idle goods according to claim 5, characterized in that, The credit scoring module (3) includes a credit data collection unit (31) and a credit analysis unit (32). The credit data collection unit (31) is used to collect users' credit data and preprocess and dimensionless the collected data, including default rate K, user rating value L, number of reports M, number of transactions completed N, and positive review rate O. The credit analysis unit (32) is used to couple the processed data, input the processed multiple data into a pre-trained deep learning framework, perform feature fusion through a multi-layer neural network, calculate and obtain the credit analysis coefficient, perform data analysis on the calculated credit analysis coefficient, and determine the user's credit rating based on the analysis results. The credit analysis unit (32) calculates the credit analysis coefficient using the following formula; ； In the formula: K is the default rate, L is the user rating, M is the number of reports, N is the number of completed transactions, and O is the positive review rate.

7. The campus resource processing system for leasing and group buying of idle goods according to claim 6, characterized in that, The specific analysis method of the credit analysis coefficient by the credit analysis unit (32) is as follows: when At this time, it indicates that the current user is a Level 3 credit user, which is a low credit rating; when At this time, it indicates that the current user is a level 2 credit user, with a medium credit rating; when At this time, it indicates that the current user is a Level 1 credit user, with a high level of credit.

8. The campus resource processing system for leasing and group buying of idle goods according to claim 7, characterized in that, The risk warning module (4) integrates and analyzes the analysis results of the behavior analysis module (2) and the credit scoring module (3) to determine whether there is any abnormality in the user's transaction. The specific analysis method is as follows: If the current user's behavior analysis result is Level 1 and the credit analysis result is Level 1, it means that the current user's transaction credibility is Level 1 and no monitoring or prompts are required. If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required. If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required. If the current user's behavior analysis result is level two and the credit analysis result is level one, it means that the current user's transaction credibility is level two and needs to be monitored. If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level three, which requires monitoring and notification to the other party in the transaction. If the current user's behavior analysis result is level two, and the credit analysis result is also level two, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required. If the current user's behavior analysis result is level three and the credit analysis result is level one, it means that the current user's transaction credibility is level one and no monitoring or prompts are required. If the current user's behavior analysis result is level three and the credit analysis result is level two, it means that the current user's transaction credibility is level two and needs to be monitored. If the current user's behavior analysis result is level three, and the credit analysis result is also level three, it means that the current user's transaction credibility is level four, and they cannot directly conduct transactions; third-party intervention and authentication are required.

9. The campus resource processing system for leasing and group buying of idle goods according to claim 8, characterized in that, The level adjustment module (5) integrates and analyzes the analysis results of the behavior analysis module (2), the credit scoring module (3), and the risk warning module (4), and determines the user level classification based on the analysis results, in the following specific way: If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 1, it means that the current user is a Level 1 user and belongs to a high-quality user. If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level one user and belongs to a high-quality user. If the current user's behavior analysis result is level two and the transaction credibility is level one, it means that the current user is a level two user, belonging to the medium quality user category; If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 2, it means that the current user is a Level 1 user and belongs to a high-quality user. If the current user's behavior analysis result is Level 2, and the transaction credibility is Level 2, it means that the current user is a Level 2 user, belonging to the medium-quality user category; If the current user's behavior analysis result is level three and the transaction credibility is level two, it means that the current user is a level three user, belonging to a low-quality user. If the current user's behavior analysis result is Level 1 and the transaction credibility is Level 3, it means that the current user is a Level 3 user, which is a low-quality user. If the current user's behavior analysis result is level two and the transaction credibility is level three, it means that the current user is a special user and needs to be specially marked. If the current user's behavior analysis result is level three, and the transaction credibility is also level three, it means that the current user is a special user and needs to be specially marked.

Images