Security house application qualification auditing method and device and readable storage medium

Abstract

The invention provides a guarantee house application qualification auditing method and device and a readable storage medium. The method comprises the steps of obtaining applicant data and evaluation rule data used for performing guarantee house application qualification auditing; performing intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result; performing three-stage evaluation based on the mapping result to obtain an evaluation result; and obtaining a guarantee house application qualification checking result of the applicant according to the evaluation result. According to the security house application qualification auditing method provided by the invention, the automation level, accuracy and efficiency of security house application qualification evaluation can be remarkably improved, and meanwhile, the fairness and transparency of the evaluation process are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a method and device for auditing the eligibility of an applicant for a welfare house, and a readable storage medium. BACKGROUND

[0002] With the continuous improvement of China's welfare housing policy and the increasing number of applicants, the auditing of the eligibility of an applicant for a welfare house is facing increasingly severe challenges.

[0003] The prior art usually uses a relational database to compare or a simple rule engine to match. Although it can perform basic rule comparison, it has obvious shortcomings in processing semantic understanding and complex condition judgment, and cannot effectively handle cases where the semantics are equivalent but the expressions are different, such as "the applicant has a college education" and "the applicant has a bachelor's degree". At the same time, most existing systems need to manually configure field mapping relationships, and lack flexibility when facing different data formats.

[0004] Therefore, how to more efficiently and reliably audit the eligibility of an applicant for a welfare house has become a problem to be solved. SUMMARY

[0005] The technical problem to be solved by the present application is to overcome the above-mentioned shortcomings of the prior art, and to provide a method and device for auditing the eligibility of an applicant for a welfare house, and a readable storage medium, to solve the problems existing in the prior art.

[0006] In a first aspect, the present application provides a method for auditing the eligibility of an applicant for a welfare house, comprising:

[0007] S1, obtaining applicant data and evaluation rule data for auditing the eligibility of an applicant for a welfare house;

[0008] S2, performing intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result;

[0009] S3, performing three-stage evaluation based on the mapping result to obtain an evaluation result;

[0010] S4, obtaining an eligibility auditing result of an applicant for a welfare house according to the evaluation result.

[0011] In some embodiments, S2 comprises:

[0012] According to the applicant data and the evaluation rule data, a large language model based on a Transformer architecture is used to analyze the semantic relationship between the evaluation rule fields and the applicant fields, and to establish a corresponding mapping between the fields.

[0013] In some embodiments, according to the applicant data and the evaluation rule data, a semantic relationship between the evaluation rule fields and the applicant fields is analyzed by a large language model based on a Transformer architecture, a corresponding mapping between the fields is established, and the method comprises the following steps:

[0014] S21, extracting a set of evaluation rule fields in the evaluation rule data;

[0015] S22, extracting a set of applicant fields in the applicant data;

[0016] S23, constructing a semantic mapping prompt word, the semantic mapping prompt word comprising the set of evaluation rule fields, the set of applicant fields, and a mapping task description;

[0017] S24, based on the semantic mapping prompt word, calling a large language model based on a Transformer architecture for semantic analysis to generate a field mapping relationship;

[0018] S25, parsing the model return result to construct a field mapping dictionary.

[0019] In some embodiments, S3 comprises:

[0020] First stage: based on accurate text matching and conditional comparison, filtering rules that do not meet the conditions;

[0021] Second stage: intelligent matching of part of the key fields through semantic analysis, further filtering rules;

[0022] Third stage: semantic understanding and comprehensive judgment of the remaining rules to obtain the final evaluation result.

[0023] In some embodiments, based on accurate text matching and conditional comparison, filtering rules that do not meet the conditions comprises:

[0024] Extracting basic information of the applicant;

[0025] Performing a simple conditional check on each rule:

[0026] Filtering rules that obviously do not meet the conditions;

[0027] Marking complex conditional rules that need detailed analysis.

[0028] In some embodiments, intelligent matching of part of the key fields through semantic analysis, further filtering rules comprises:

[0029] Extracting key fields of the rules;

[0030] Constructing an intermediate screening prompt word, the intermediate screening prompt word comprising simplified rule data and applicant data;

[0031] Call large language model for semantic analysis, process simple semantic matching;

[0032] Parse the return result of the large language model, and further filter the rules that do not meet the conditions.

[0033] In some embodiments, the remaining rules are subjected to semantic understanding and comprehensive judgment to obtain the final evaluation result, including:

[0034] Construct a detailed analysis prompt word, which contains complete rule data, applicant data and field mapping relationship;

[0035] Call large language model for deep semantic understanding and evaluate all conditions;

[0036] Parse the return result of the large language model to determine the final rule that meets the conditions;

[0037] Generate a detailed analysis report and evaluation conclusion.

[0038] In some embodiments, S4, including:

[0039] Structurally summarize the evaluation results to obtain standardized evaluation output;

[0040] Among them, the evaluation output content includes:

[0041] Whether it meets the application conditions;

[0042] Specific reason for compliance or non-compliance;

[0043] Process record of each evaluation stage;

[0044] The final matching policy rule and its corresponding condition list.

[0045] In a second aspect, the application provides a security housing application qualification review device, the device comprises:

[0046] A data acquisition module is configured to acquire applicant data and evaluation rule data for security housing application qualification review;

[0047] A field mapping module is configured to perform intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result;

[0048] A three-stage evaluation module is configured to perform three-stage evaluation based on the mapping result to obtain an evaluation result;

[0049] A result determination module is configured to obtain a security housing application qualification review result of the applicant according to the evaluation result.

[0050] In a third aspect, the present application provides a device for auditing the qualification of an applicant for a welfare house, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for auditing the qualification of an applicant for a welfare house according to the first aspect.

[0051] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method for auditing the qualification of an applicant for a welfare house according to the first aspect.

[0052] The method for auditing the qualification of an applicant for a welfare house, the device and the computer readable storage medium provided by the present application comprise the following steps: obtaining applicant data and evaluation rule data for auditing the qualification of an applicant for a welfare house; performing intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result; performing three-stage evaluation based on the mapping result to obtain an evaluation result; and obtaining an auditing result of the qualification of an applicant for a welfare house according to the evaluation result. The method for auditing the qualification of an applicant for a welfare house provided by the present application can significantly improve the automation level, accuracy and efficiency of the evaluation of the qualification of an applicant for a welfare house, while ensuring the fairness and transparency of the evaluation process. BRIEF DESCRIPTION OF DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0054] Figure 1 A flowchart of a method for auditing the qualification of an applicant for a welfare house according to an embodiment of the present application;

[0055] Figure 2 A flowchart of another method for auditing the qualification of an applicant for a welfare house according to an embodiment of the present application;

[0056] Figure 3 A schematic diagram of the intelligent field mapping preprocessing of the applicant data and the evaluation rule data according to an embodiment of the present application;

[0057] Figure 4 A schematic diagram of the three-stage evaluation process according to an embodiment of the present application;

[0058] Figure 5 A structural schematic diagram of a device for auditing the qualification of an applicant for a welfare house according to an embodiment of the present application;

[0059] Figure 6 A structural schematic diagram of another device for auditing the qualification of an applicant for a welfare house according to an embodiment of the present application.

[0060] The specific embodiments of the present application have been shown through the above drawings, and will be described in more detail hereinafter. The drawings and the written description are not intended to limit the scope of the present application in any way, but to explain the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0061] In order for those skilled in the art to better understand the technical solutions of the present application, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0062] It can be understood that the specific embodiments and drawings described herein are only used to explain the present application, but not to limit the present application.

[0063] It can be understood that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0064] It can be understood that, for the purpose of description, only parts related to the present application are shown in the drawings of the present application, and parts unrelated to the present application are not shown in the drawings.

[0065] It can be understood that each unit and module involved in the embodiments of the present application can only correspond to one entity structure, or can be composed of multiple entity structures, or multiple units and modules can be integrated into one entity structure.

[0066] It can be understood that the terms "first", "second" and the like in the embodiments of the present application are used to distinguish different objects, or to distinguish different treatments of the same object, but not to describe a specific order of the objects.

[0067] It can be understood that the functions and steps marked in the flowcharts and block diagrams of the present application can occur in an order different from that marked in the drawings without conflict.

[0068] It can be understood that in the flowcharts and block diagrams of the present application, the system, device, equipment, method according to the embodiments of the present application are shown as possible implementation architecture, function and operation. Each block in the flowchart or block diagram can represent a unit, module, program segment, code, which contains executable instructions for realizing the specified function. Moreover, each block or combination of blocks in the block diagram and flowchart can be realized by a hardware-based system for realizing the specified function, or by a combination of hardware and computer instructions.

[0069] It can be understood that the units and modules involved in the embodiments of the present application can be realized in the form of software or hardware, for example, the units and modules can be located in a processor.

[0070] It can be understood that the specific values of the parameters in the present application are only exemplary, and in actual application, parameter optimization adjustment processing can be performed based on specific requirements.

[0071] Currently, the technical problems mainly exist in the following aspects in the eligibility review mode of the security housing application:

[0072] 1. Data format diversity problem: The applicant data format and evaluation rules of each region are significantly different, including non-uniform field naming, data structure difference, etc., which leads to difficulty in automatic review. For example, the field representing the marital status may be named "marital status", "marital situation" or "marital relationship" in different systems.

[0073] 2. Policy rule complexity problem: The security housing related policy rules usually involve multiple dimensions of evaluation conditions, including explicit hard indicators such as age and income, and fuzzy conditions such as "having corresponding professional qualifications" and "belonging to key talents" that rely on semantic understanding. Due to the complex structure and various forms of expression of such rules, traditional rule engines based on fixed logic have the problem of insufficient accuracy in handling the above scenarios, and lack flexibility in dealing with frequent policy adjustments and regional differences.

[0074] 3. Artificial review efficiency bottleneck: The existing artificial review method is time-consuming and easily affected by subjective factors, making it difficult to ensure the consistency of evaluation standards.

[0075] The existing technology usually adopts a relational database comparison or a simple rule engine for matching, which can perform basic rule comparison, but has obvious shortcomings in handling semantic understanding and complex condition judgment, and cannot effectively handle cases such as "applicants with college education" and "with bachelor's degree" that are semantically equivalent but expressed differently. At the same time, most existing systems need to manually configure field mapping relationships, and lack flexibility when facing different data formats.

[0076] In addition, the existing security housing qualification evaluation system mostly uses rule engines or fixed logic to match applicant information and policy rules. Although it can handle some structured rules, it has obvious shortcomings in dealing with rule expression diversity, complex conditions, data heterogeneity and frequent policy adjustments, and cannot meet the requirements of intelligence, efficiency and explainability in actual application. Especially in the face of semantic ambiguity and non-uniform fields, traditional methods have low processing efficiency, high error rate and rely on a lot of manual intervention, which limits the performance and expansion capability of the system.

[0077] Based on the problems existing in the prior art, the present application proposes a security housing application qualification evaluation method based on intelligent field mapping and multi-stage semantic analysis, and the main technical ideas of the present application include:

[0078] 1. Intelligent field mapping: Utilize large language models to achieve automatic semantic mapping between rule fields and applicant fields, solving data heterogeneity problems.

[0079] 2. Three-stage progressive evaluation mechanism:

[0080] First stage: Quick screening, based on precise field matching, filtering obvious rules that do not meet the conditions

[0081] Second stage: Intermediate screening, preliminary matching through semantic analysis

[0082] Third stage: Detailed analysis, using deep semantic understanding to handle complex conditions

[0083] 3. Dynamic semantic understanding framework: Build a special policy rule semantic understanding framework to support rule recognition in different forms of expression.

[0084] The present application proposes a method for evaluating the eligibility of affordable housing applicants, with stronger semantic understanding ability, self-adaptability and evaluation efficiency. The specific technical purposes are as follows:

[0085] 1. For the problem of manual configuration of field mapping and inability to automatically process data heterogeneity in the prior art, the present application aims to provide an intelligent field mapping method based on large language models, to realize automatic semantic mapping between applicant data and evaluation rules, and to improve system adaptability.

[0086] 2. For the problem that the rule engine in the prior art can only handle simple Boolean logic and numerical comparison, and cannot understand complex semantic conditions, the present application aims to establish a multi-stage semantic analysis mechanism to handle evaluation conditions of different complexity in layers, and to improve the semantic understanding ability of the system.

[0087] 3. For the problem of fixed evaluation process and low processing efficiency in the prior art, the present application aims to realize a three-stage progressive evaluation method, to optimize evaluation efficiency through the combination of quick screening, intermediate screening and detailed analysis.

[0088] 4. For the problem of poor system scalability and difficulty in adapting to policy changes in the prior art, the present application aims to build a dynamically configurable evaluation framework to support flexible access and update of different regional policy rules.

[0089] 5. For the problem of lack of explainability of evaluation results in the prior art, the present application aims to provide detailed evaluation process tracking and result explanation mechanism to improve the credibility and traceability of evaluation results.

[0090] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes can not be described again in some examples. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0091] The present application provides a method for auditing the eligibility of a housing application. The working process of the method can be realized by electronic devices such as computers, handheld smart terminals, etc. For ease of explanation, the method execution subject in the embodiments of the present application is a computer.

[0092] Figure 1 A schematic diagram of the method for auditing the eligibility of a housing application provided by the embodiments of the present application is shown in Figure 2 A schematic diagram of the method for auditing the eligibility of a housing application provided by the embodiments of the present application is shown in Figure 1 and Figure 2 The present application provides a method for auditing the eligibility of a housing application, which includes S1-S4, and the details are as follows:

[0093] S1, obtaining applicant data and evaluation rule data for auditing the eligibility of a housing application;

[0094] The applicant data and the evaluation rule data are both in JSON format. If they are in other formats, the data will be converted from the other formats to the JSON format.

[0095] S2, performing intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result;

[0096] In some embodiments, S2 includes:

[0097] According to the applicant data and the evaluation rule data, a large language model based on a Transformer architecture is used to analyze the semantic relationship between the evaluation rule fields and the applicant fields, and a corresponding mapping between the fields is established.

[0098] Figure 3 A schematic diagram of the intelligent field mapping preprocessing of the applicant data and the evaluation rule data provided by the embodiments of the present application is shown in Figure 3 In some embodiments, according to the applicant data and the evaluation rule data, a large language model based on a Transformer architecture is used to analyze the semantic relationship between the evaluation rule fields and the applicant fields, and a corresponding mapping between the fields is established, including:

[0099] S21, extracting an evaluation rule field set rule_fields from the evaluation rule data;

[0100] S22, extract the applicant field set in the applicant data;

[0101] S23, construct a semantic mapping prompt word, which contains an evaluation rule field set, an applicant field set, and a mapping task description;

[0102] S24, based on the semantic mapping prompt word, call the large language model based on the Transformer architecture for semantic analysis, and generate the field mapping relationship;

[0103] S25, parse the model return result, and construct a field mapping dictionary mapping.

[0104] Among them, the semantic mapping prompt word template in S24 is as follows:

[0105] ###

[0106] Analyze the following two groups of fields to establish their semantic mapping relationship:

[0107] Evaluation rule field: [field1, field2,...]

[0108] Applicant field: [field A, field B,...]

[0109] Return a JSON mapping with the following format:

[0110]

[0111] ###

[0112] When the result returned by the large language model cannot be correctly parsed into JSON format, the system will try to correct the format or fall back to the basic mapping relationship to ensure the stability of the mapping process. The temperature parameter is set to 0.1-0.2, which is based on system test results. In this range, the model output has sufficient certainty and consistency, while maintaining the necessary flexibility of semantic understanding.

[0113] S3, based on the mapping result, three-stage evaluation is carried out, and the evaluation result is obtained;

[0114] In some embodiments, S3 includes:

[0115] First stage: based on accurate text matching and conditional comparison, filter rules that do not meet the conditions;

[0116] Second stage: through semantic analysis, intelligent matching is carried out on part of the key fields, and the rules are further filtered;

[0117] Third stage: semantic understanding and comprehensive judgment are carried out on the remaining rules to obtain the final evaluation result.

[0118] Figure 4 A schematic diagram of the three-stage evaluation process provided by the embodiments of the present application is shown in FIG. 1, in some embodiments, based on precise text matching and conditional comparison, rules that do not meet the conditions are filtered, including: Figure 4

[0119] Extracting the basic information of the applicant;

[0120] Performing a simple conditional check on each rule:

[0121] Filtering rules that obviously do not meet the conditions;

[0122] Marking complex conditional rules that require detailed analysis.

[0123] Specifically, the first stage is rapid screening. In this stage, based on precise text matching and conditional comparison, the rules that obviously do not meet the conditions are quickly filtered. The specific implementation process includes:

[0124] 1. Extracting the basic information of the applicant (marital status, age, etc.);

[0125] 2. Performing a simple conditional check on each rule:

[0126] (1) When a specific limit is not set for a field in the rule (for example, the field value is “ / ” or empty), it is considered that the field does not constitute a limiting condition in this rule, and it is automatically determined to be in compliance;

[0127] (2) Performing direct matching or numerical comparison operations on fields such as marital status and age;

[0128] 3. Filtering out rules that obviously do not meet the conditions;

[0129] 4. Marking complex conditional rules that require detailed analysis.

[0130] In this stage, the system mainly performs precise text matching, for example, “married” only matches “married” and does not perform semantic expansion. For age conditions, numerical comparison is performed.

[0131] In some embodiments, intelligent matching is performed on part of the key fields through semantic analysis to further screen the rules, including:

[0132] Extracting the key fields of the rules;

[0133] Constructing intermediate screening cue words, which contain simplified rule data and applicant data;

[0134] Calling a large language model to perform semantic analysis and handle simple semantic matching;

[0135] ​The returned result of the large language model is analyzed to further filter rules that do not meet the conditions.

[0136] Specifically, the second stage is intermediate screening. In this stage, the key fields are intelligently matched through semantic analysis to further screen the rules. The specific implementation process includes:

[0137] 1. Extract the key fields of the rules (family type, applicant type, residence, social security, etc.);

[0138] 2. Build intermediate screening prompt words, including simplified rule data and applicant data;

[0139] 3. Call the language model for semantic analysis, focusing on simple semantic matching;

[0140] 4. Analyze the returned result to further filter rules that do not meet the conditions.

[0141] This stage mainly handles simple conditions that require semantic understanding, such as identifying "married", "married", "with spouse" as the same semantic, or determining whether "college education" meets the requirements of "college education and above".

[0142] In some embodiments, semantic understanding and comprehensive judgment are performed on the remaining rules to obtain the final evaluation result, including:

[0143] Build detailed analysis prompt words, which include complete rule data, applicant data, and field mapping relationships;

[0144] Call the large language model for deep semantic understanding and evaluate all conditions;

[0145] Analyze the returned result of the large language model to determine the final rules that meet the conditions;

[0146] Generate a detailed analysis report and evaluation conclusion.

[0147] Specifically, the third stage is detailed analysis. In this stage, the remaining rules are comprehensively and deeply understood and comprehensively judged to obtain the final evaluation result. The specific implementation process includes:

[0148] 1. Build detailed analysis prompt words, including complete rule data, applicant data, and field mapping relationships;

[0149] 2. Call the language model for deep semantic understanding to comprehensively evaluate all conditions;

[0150] 3. Analyze the returned result to determine the final rules that meet the conditions;

[0151] 4. Generate a detailed analysis report and evaluation conclusion.

[0152] The third stage handles the most complex conditional judgments, including policy requirements with long text descriptions, multi-field combined conditions, and the like, and relies on the advanced understanding and reasoning capabilities of the language model to achieve accurate evaluation.

[0153] The present application uses a three-stage evaluation process, combining precise matching and semantic understanding techniques, to sequentially filter and analyze policy rules from shallow to deep, ensuring accuracy while improving evaluation efficiency and system adaptability.

[0154] S4. Obtain the applicant's affordable housing application qualification review result based on the evaluation result.

[0155] In some embodiments, S4 includes:

[0156] Structurally summarize the evaluation results to obtain standardized evaluation output;

[0157] The evaluation output content includes:

[0158] Whether it meets the application conditions;

[0159] Specific reason for compliance or non-compliance;

[0160] Process records of each evaluation stage;

[0161] The final matching policy rules and their corresponding condition list.

[0162] Specifically, this step structurally summarizes the evaluation results of the three-stage evaluation process to form standardized evaluation output, supporting system integration, audit traceability, and subsequent use. The specific implementation process includes:

[0163] 1. Whether it meets the application conditions (represented in Boolean form);

[0164] 2. Specific reason for compliance or non-compliance, providing explainable evaluation basis;

[0165] 3. Process records of each evaluation stage, detailing the judgment logic and processing results at each step;

[0166] 4. The final matching policy rules and their corresponding condition list.

[0167] The evaluation results are stored in JSON format for easy system integration and subsequent processing. At the same time, the system saves key evaluation records in the database, supporting audit traceability. For evaluation results that require special protection, the system supports generating evaluation proofs with digital signatures, ensuring the reliability and tamper resistance of the evaluation results. The system also supports multi-language policy rule parsing, capable of handling policy rules in different regions and languages, greatly enhancing the adaptability and scalability of the system.

[0168] The key technical points and protection points of the present application include:

[0169] 1. Intelligent field mapping technology

[0170] (1) Innovation point: Utilize large language models to achieve automatic semantic mapping of rule fields and applicant fields;

[0171] (2) Technical features:

[0172] Use specific mapping prompt word templates to guide the model to understand the semantic relationship between fields;

[0173] Set a low temperature parameter of 0.1-0.2 to ensure the consistency of the mapping results;

[0174] Include a multi-layer exception handling mechanism that can automatically fall back to basic mapping when mapping fails;

[0175] (3) Protection value: Solves the technical bottleneck of manual configuration of field mapping in traditional systems.

[0176] 2. Three-stage progressive evaluation mechanism

[0177] (1) Innovation point: Divide the evaluation process into three progressive stages: quick screening, intermediate screening, and detailed analysis;

[0178] (2) Technical features:

[0179] Use precise matching algorithms in the quick screening stage to filter out rules that obviously do not meet the conditions;

[0180] Introduce semantic understanding in the intermediate screening stage to handle simple semantic equivalence judgments;

[0181] Use deep semantic understanding in the detailed analysis stage to handle complex policy conditions;

[0182] (3) Protection value: Achieves the optimal balance between evaluation efficiency and accuracy.

[0183] 3. Dynamic semantic understanding framework

[0184] (1) Innovation point: Constructed a semantic understanding framework specifically for affordable housing policy rules;

[0185] (2) Technical features:

[0186] Design specific prompt word templates to improve the model's understanding accuracy of policy conditions;

[0187] Implement a dynamic adaptation mechanism for condition analysis to support rule recognition in different expression modes;

[0188] Establish a semantic equivalence judgment mechanism for rule conditions;

[0189] (3) Protection value: Breaks the limitation of traditional rule engine that can only handle fixed format rules.

[0190] 4. Multi-level exception handling mechanism

[0191] (1) Innovation point: Designed a multi-level fault-tolerant processing scheme for semantic analysis failure;

[0192] (2) Technical features:

[0193] Realize the reliability verification of evaluation results;

[0194] Support automatic rollback to the previous stage result;

[0195] Include preset alternative processing flow;

[0196] (3) Protection value: Ensure the stability of the system in complex scenarios.

[0197] 5. Traceable evaluation result generation mechanism

[0198] (1) Innovation point: Realize the complete evaluation process record and result explanation function;

[0199] (2) Technical features:

[0200] Record the screening process and basis of each evaluation stage;

[0201] Generate detailed description of rule matching;

[0202] Support digital signature verification of evaluation results;

[0203] (3) Protection value: Provide credibility guarantee for evaluation results.

[0204] The application provides a method for guaranteeing housing application qualification review, which can significantly improve the automation level, accuracy and efficiency of housing application qualification evaluation, while ensuring the fairness and transparency of the evaluation process.

[0205] It should be understood that although each step in the flowchart in the above embodiment is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps has no strict order limitation, and they can be executed in other orders. Moreover, at least part of the steps in the figure can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order is not necessarily sequential, but can be alternately executed with other steps or sub-steps or stages of other steps.

[0206] Figure 5 A schematic diagram of a security housing application qualification review device provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the present application provides a security housing application qualification review device, which comprises: Figure 5 A data acquisition module 11 configured to acquire applicant data and evaluation rule data for security housing application qualification review;

[0207] A field mapping module 12 configured to perform intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain a mapping result;

[0208] A three-stage evaluation module 13 configured to perform three-stage evaluation based on the mapping result to obtain an evaluation result;

[0209] A result determination module 14 configured to obtain a security housing application qualification review result of the applicant according to the evaluation result.

[0210] For the limitation of the security housing application qualification review device, reference can be made to the limitation of the security housing application qualification review method in the above-mentioned embodiments of the present application, which will not be repeated here in the present embodiment.

[0211]

[0212] Another schematic diagram of a security housing application qualification review device provided by an embodiment of the present application is shown in FIG. 2. As shown in FIG. 2, the device comprises a memory 22 and a processor 21, the memory stores a computer program, and the processor is configured to run the computer program to execute the method in the above-mentioned embodiments of the present application. Figure 6 Figure 6 The memory is connected with the processor, and the memory can adopt a flash memory or a read-only memory or other memories, and the processor can adopt a central processing unit or a single-chip microcomputer.

[0213] In some embodiments, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method in the above-mentioned embodiments of the present application.

[0214]

[0215] ​​The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules, or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), DVD or other optical disc storage, cartridges, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer.

[0216] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of this application, and this application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this application, and these modifications and improvements are also considered to be within the scope of protection of this application.

Claims

1. A method for verifying eligibility for affordable housing applications, characterized in that, The method includes: S1. Obtain applicant data and assessment rule data used for eligibility review of affordable housing applications; S2. Perform intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain the mapping result; S3. Perform a three-stage evaluation based on the mapping results to obtain the evaluation results; S4. Based on the assessment results, obtain the applicant's eligibility review results for subsidized housing.

2. The method for verifying eligibility for affordable housing applications according to claim 1, characterized in that, S2 includes: Based on the applicant data and the evaluation rule data, the semantic relationship between the evaluation rule fields and the applicant fields is analyzed using a large language model based on the Transformer architecture, and a corresponding mapping between the fields is established.

3. The method for verifying eligibility for affordable housing applications according to claim 2, characterized in that, Based on the applicant data and the assessment rule data, a large language model based on the Transformer architecture is used to analyze the semantic relationship between the assessment rule fields and the applicant fields, and to establish a mapping between the fields, including: S21. Extract the set of evaluation rule fields from the evaluation rule data; S22. Extract the set of applicant fields from the applicant data; S23. Construct semantic mapping prompts, wherein the semantic mapping prompts include a set of evaluation rule fields, a set of applicant fields, and a description of the mapping task; S24. Based on semantic mapping prompts, call a large language model based on the Transformer architecture to perform semantic analysis and generate field mapping relationships; S25. Parse the model's returned results and construct a field mapping dictionary.

4. The method for verifying eligibility for affordable housing applications according to claim 1, characterized in that, S3 includes: Phase 1: Filtering rules that do not meet the conditions based on precise text matching and condition comparison; Phase Two: Intelligent matching of some key fields through semantic analysis to further refine the filtering rules; The third stage involves semantic understanding and comprehensive judgment of the remaining rules to obtain the final evaluation result.

5. The method for verifying eligibility for affordable housing applications according to claim 4, characterized in that, Based on precise text matching and conditional comparison, rules that do not meet the conditions are filtered out, including: Extract the applicant's basic information; Perform a simple condition check on each rule: Filter rules that clearly do not meet the criteria; Mark complex conditional rules that require detailed analysis.

6. The method for verifying eligibility for affordable housing applications according to claim 4, characterized in that, Semantic analysis is used to intelligently match certain key fields, and further filtering rules are implemented, including: Key fields for extraction rules; Construct intermediate-level filtering suggestions, which include simplified rule data and applicant data; Call upon a large language model to perform semantic analysis and handle simple semantic matching; The results returned by the large language model are analyzed to further filter out rules that do not meet the conditions.

7. The method for verifying eligibility for affordable housing applications according to claim 4, characterized in that, The remaining rules are semantically understood and comprehensively judged to obtain the final evaluation result, including: Construct detailed analysis prompts, which include complete rule data, applicant data, and field mapping relationships; Call upon a large language model for deep semantic understanding and evaluate all conditions; Analyze the return results of the large language model to determine the final rules that meet the conditions; Generate detailed analysis reports and evaluation conclusions.

8. The method for verifying eligibility for affordable housing applications according to any one of claims 1-7, characterized in that, S4 includes: The evaluation results are then structured and summarized to obtain standardized evaluation output. The evaluation output includes: Do you meet the application requirements? A detailed explanation of whether the application meets or does not meet the requirements; Process records for each evaluation stage; The final list of matching policy rules and their corresponding conditions.

9. A device for verifying eligibility for affordable housing applications, characterized in that, The device includes: The data acquisition module is set up to acquire applicant data and assessment rule data used for eligibility review of affordable housing applications; The field mapping module is configured to perform intelligent field mapping preprocessing on the applicant data and the evaluation rule data to obtain mapping results; The three-stage evaluation module is configured to perform a three-stage evaluation based on the mapping results to obtain the evaluation results. The result determination module is configured to obtain the applicant's eligibility review result for subsidized housing based on the assessment results.

10. A device for verifying eligibility for affordable housing applications, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to implement the method for reviewing eligibility for affordable housing applications as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for verifying eligibility for affordable housing applications as described in any one of claims 1-8.

Images