Method and system for generating executive pedestrian performance ability report based on large model

Abstract

The invention provides a method and a system for generating a performance ability report of a licensed pedestrian based on a large model, electronic equipment and a storage medium, and aims to solve the problem that a result may be misjudged due to the fact that a high-consumption record only depends on simple keyword matching, the method comprises the following steps: uploading payment detail data of the licensed pedestrian, and performing data cleaning; according to the case judgment file, extracting case key information including a file number, a case target, a licensed pedestrian name and a destination; inputting the cleaned payment detail data, understanding the consumption type and the consumption place by the large model, and outputting a consumption result; the output consumption result is matched with the target place and law and regulation information of the permitted pedestrian, and the high consumption behavior of the permitted pedestrian and the consumption behavior necessary for non-production life are judged; and a permitted pedestrian performance ability report is generated accordingly. According to the method and the device, the aim of generating the personal performance ability report more intelligently and more accurately can be achieved, and the efficiency of legal work is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and in particular to a method for generating a report on the performance ability of a person subject to execution based on a large model, a system for generating a report on the performance ability of a person subject to execution based on a large model, an electronic device, and a computer-readable storage medium. Background Art

[0002] At present, in the case where the person subject to execution refuses to perform the court judgment, the consumption details data provided by the mobile payment provider are usually uploaded to the system. The system will screen out high-consumption data according to a unified logic. The judge will check the statistical results and manually define whether the person subject to execution has the ability to perform the contract based on the subject matter of the case and the relevant laws and regulations of the target location of the person subject to execution. However, because fields such as consumption type and consumption location involve commercial secrets and personal privacy, mobile payment providers do not provide this information to the system. The system can only use mapping relationships to achieve matching queries between consumption names and consumption types and consumption locations. This traditional keyword matching method requires a large number of enumerations of mapping relationships, and there is a situation where statistical results are omitted due to incomplete enumeration, which in turn affects the statistical results and the judge's judgment.

[0003] The current method of generating the report on the performance ability of the person subject to execution only relies on simple keyword matching. The matching results are highly dependent on the comprehensiveness and accuracy of the mapping relationship table maintenance. Incomplete maintenance of the mapping relationship may lead to missing out on truly valuable details closely related to high consumption, resulting in misjudgment of the results. At the same time, the current system defines high consumption using fixed logic, and fails to make flexible individual judgments based on the case and the registered residence / activity location of the person subject to execution. Under the "one-size-fits-all" judgment logic, when manually defining the conclusion, it is necessary to manually check the details and do data processing to draw a conclusion. Summary of the invention

[0004] In order to at least solve the problem in the prior art that high-consumption records fail to make flexible and personalized judgments based on the cases and the domicile / activity places of the persons subject to execution, and the matching query of consumption names, consumption types and consumption locations only relies on simple keyword matching, which may lead to misjudgment of the results, the present disclosure provides a method for generating a report on the performance ability of the persons subject to execution based on a big model, a system for generating a report on the performance ability of the persons subject to execution based on a big model, an electronic device and a computer-readable storage medium, which can achieve the goal of generating personal performance ability reports more intelligently and accurately and improve the efficiency of legal work.

[0005] In a first aspect, the present disclosure provides a method for generating a report on the performance ability of a person subject to execution based on a large model, the method comprising:

[0006] Upload the payment details of the person subject to execution and clean the data;

[0007] Extract key case information based on the case judgment documents, including the case file number, case subject, name of the person to be executed, and target location;

[0008] Input payment details after data cleaning, and the big model will understand the consumption type and location, and output the consumption results;

[0009] Match the output consumption results with the target location and legal and regulatory information of the person subject to execution, and determine the person's high-consumption behavior and consumption behavior that is not necessary for production and life;

[0010] A report on the debtor's ability to perform is generated based on the debtor's high consumption behavior and consumption behavior that is not necessary for production and life.

[0011] Furthermore, the method further comprises:

[0012] Input documents related to minimum living security, use NLP (Natural Language Processing) to refine the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the refined location text, year, and minimum living security standard numbers into the knowledge base.

[0013] Furthermore, the payment details data after the input data cleaning is understood by the big model based on the consumption type and consumption location, and the consumption results are output, including:

[0014] According to the business scenario of the performance ability report of the person subject to execution, a prompt project is formulated for obtaining consumption type and consumption location respectively;

[0015] Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions;

[0016] After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

[0017] Furthermore, the method further comprises:

[0018] Develop prompt projects for obtaining property clues;

[0019] When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person subject to execution.

[0020] Furthermore,

[0021] The output consumption results are matched with the target location and legal and regulatory information of the person subject to execution, and the high-consumption behaviors and consumption behaviors that are not necessary for production and life of the person subject to execution are determined, including:

[0022] Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination;

[0023] For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base;

[0024] The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location;

[0025] The report on the ability of the person subject to execution to generate the report based on the person subject to execution’s high-consumption behavior and non-productive and living consumption behavior includes:

[0026] Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

[0027] Furthermore, the method further comprises:

[0028] According to actual conditions, adjust the report data statistical time period and adapt local laws and regulations information, and regenerate the report.

[0029] In a second aspect, the present disclosure provides a system for generating a report on the performance ability of a person subject to execution based on a large model, the system comprising:

[0030] The upload module is configured to upload payment details of the person subject to execution and perform data cleansing;

[0031] An extraction module is configured to extract key case information, including file number, case subject, name of the person to be executed, and target location, based on the case judgment document;

[0032] The data processing module is set to input the payment details after data cleaning, and the big model understands the consumption type and consumption location, and outputs the consumption results;

[0033] A determination module is configured to match the output consumption results with the destination of the person subject to execution and the legal and regulatory information, and determine the high-consumption behavior and consumption behavior that is not necessary for production and life of the person subject to execution;

[0034] A generation module is configured to generate a report on the performance ability of the person subject to execution based on the person's high consumption behavior and consumption behavior that is not necessary for production and life.

[0035] Furthermore, the system also includes:

[0036] The extraction module is set to input documents related to the minimum living security, use NLP to extract the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the extracted location text, year, and minimum living security standard numbers into the knowledge base.

[0037] Furthermore, the data processing module is specifically configured as follows:

[0038] According to the business scenario of the performance ability report of the person subject to execution, a prompt project is formulated for obtaining consumption type and consumption location respectively;

[0039] Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions;

[0040] After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

[0041] Furthermore, the data processing module is also configured to:

[0042] Develop prompt projects for obtaining property clues;

[0043] When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person subject to execution.

[0044] Furthermore,

[0045] The determination module is specifically configured as follows:

[0046] Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination;

[0047] For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base;

[0048] The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location;

[0049] The generation module is specifically configured as follows:

[0050] Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

[0051] Furthermore, the system also includes an update module;

[0052] The updating module is configured to adjust the statistical time period of the report data and adapt the local laws and regulations information according to the actual situation, so that the generating module regenerates the report.

[0053] In a third aspect, the present disclosure provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for generating a report on the performance ability of the person subject to execution based on a large model as described in any one of the first aspects.

[0054] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for generating a report on the performance ability of the person subject to execution based on a large model as described in any one of the above-mentioned first aspects.

[0055] Beneficial effects:

[0056] The method for generating a report on the ability of the person to be executed based on a big model, the system for generating a report on the ability of the person to be executed based on a big model, the electronic device and the storage medium provided by the present disclosure use the big model to understand the payment details of the person to be executed, analyze the type of consumption, location of consumption and other information, and solve the problem of inaccurate / omission matching results caused by incomplete maintenance of mapping relationships. At the same time, it supports flexible screening of high-consumption data based on the laws and regulations of the target location of the person to be executed in individual cases, and the big model generates a report conclusion based on the high-consumption judgment results. By innovating the process of generating the report on the ability of the person to be executed, the goal of generating a more intelligent and accurate personal performance report can be achieved, which is of great significance to improving the efficiency of legal work. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A flowchart of a method for generating a report on the performance capability of a person subject to execution based on a large model provided in the first embodiment of the present disclosure;

[0058] Figure 2 A schematic diagram of a legal and regulatory knowledge base construction process provided in Embodiment 2 of the present disclosure;

[0059] Figure 3A schematic diagram of a case information database construction process provided in the second embodiment of the present disclosure;

[0060] Figure 4 A schematic diagram of a process for generating a report on the performance ability of a person subject to execution provided in the second embodiment of the present disclosure;

[0061] Figure 5 This is an architecture diagram of a system for generating a report on the performance capability of a person subject to execution based on a large model provided in Embodiment 3 of the present disclosure;

[0062] Figure 6 This is an architecture diagram of an electronic device provided in Embodiment 4 of the present disclosure. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention, rather than to limit the present invention.

[0064] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence; and, in the absence of conflict, the embodiments in the present disclosure and the features in the embodiments can be arbitrarily combined with each other.

[0065] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. The singular forms of "a", "said" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings.

[0066] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present disclosure, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0067] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned problems existing in the prior art are described in detail below with specific embodiments. It is understandable that in the embodiments of the present application, the execution subject can perform some or all of the steps in the embodiments of the present application, and these steps or operations are only examples. The embodiments of the present application can also perform other operations or deformations of various operations. In addition, each step can be performed in a different order presented in the embodiments of the present application, and it is possible not to perform all the operations in the embodiments of the present application. And, the following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0068] Figure 1 A flowchart of a method for generating a report on the performance ability of a person subject to execution based on a large model is provided in the first embodiment of the present disclosure, such as Figure 1 As shown, the method includes:

[0069] Step S101: Upload payment details of the person subject to execution and perform data cleaning;

[0070] Step S102: Extract key case information based on the case judgment document, including the case file number, case subject, name of the person to be executed, and target location;

[0071] Step S103: Input the payment details after data cleaning, and let the big model understand the consumption type and consumption location, and output the consumption results;

[0072] Step S104: Match the output consumption results with the destination of the person subject to execution and the legal and regulatory information to determine the high-consumption behaviors and consumption behaviors that are not necessary for production and life of the person subject to execution;

[0073] Step S105: Generate a report on the debtor's ability to perform based on the debtor's high-consumption behaviors and consumption behaviors that are not necessary for production and life.

[0074] The current method of generating the personal performance capacity report of the person subject to execution only relies on simple keyword matching. The matching result is highly dependent on the comprehensiveness and accuracy of the mapping relationship table maintenance. Incomplete maintenance of the mapping relationship may lead to missing out on truly valuable details closely related to high consumption. In addition, the definition of high consumption is divided using fixed logic, which may result in inappropriate judgments.

[0075] To solve this problem, the disclosed embodiment interacts with relevant departments and systems, such as banks, asset management platforms, and various payment centers, to obtain payment details related to the person to be executed (including mobile payment, card swiping, etc.), cleans the data, and removes the records marked with "payment failure, refund success" in the payment details, as well as the associated income / expenditure records. Then input the case judgment file, use NLP to understand and segment the case information, extract key information such as the file number, case subject, name of the person to be executed, and target location, and save it in the case information database for subsequent use. The target location is the registered permanent residence or residence of the person to be executed. The consumption details obtained after cleaning the payment details data are understood by the big model (the big language model is selected according to the actual situation) in terms of consumption type and consumption location, and the results are output; the detailed conditions of consumption type and consumption location are analyzed, and the consumption results are output; the output consumption results are matched with the target location and legal and regulatory information of the person to be executed, and the high-consumption behavior and consumption behavior that are not necessary for production and life of the person to be executed are determined; and a report on the performance ability of the person to be executed is generated based on this, including detailed records and relevant regulatory documents.

[0076] The disclosed embodiment uses a large model to understand the payment details of the person being executed, analyzes information such as consumption type and consumption location, and solves the problem of inaccurate / missing matching results due to incomplete maintenance of mapping relationships. At the same time, it supports flexible screening of high-consumption data based on the laws and regulations of the target location of the person being executed in the individual case, and the large model generates a report conclusion based on the high-consumption judgment results. By innovating the process of generating reports on the ability of the person being executed to perform, the goal of generating personal performance reports more intelligently and accurately is achieved, which is of great significance to improving the efficiency of legal work.

[0077] Furthermore, the method further comprises:

[0078] Input documents related to minimum living security, use NLP to refine the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the refined location text, year, and minimum living security standard numbers into the knowledge base.

[0079] By inputting the legal documents related to minimum living security from all over the country / the Supreme Court, NLP is used to understand and refine the language of the minimum living security standards in the laws and regulations, and the key information of the minimum living security standards in each year is extracted by location, and the extracted location text, year, and minimum living security standard numbers are saved in the legal and regulatory knowledge base for subsequent use. It can easily judge the subsequent consumer behavior and provide the corresponding judgment basis.

[0080] Furthermore, the payment details data after the input data cleaning is understood by the big model based on the consumption type and consumption location, and the consumption results are output, including:

[0081] According to the business scenario of the performance ability report of the person subject to execution, a prompt project is formulated for obtaining consumption type and consumption location respectively;

[0082] Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions;

[0083] After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

[0084] According to the business scenario of the performance report of the person subject to execution, prompt projects are formulated for obtaining consumption types and consumption locations. According to the output results, repeated debugging is carried out to finally form a prompt project with clear background information and output instructions. After entering the consumption details, the prompt project is understood based on the big model and the fixed format results are output.

[0085] Prompt is a key to unlock the model's capabilities, allowing these capabilities formed in the "pre-training" stage to be put to use; prompting the model to recall the capabilities it learned during pre-training. Prompt engineering can help users more effectively control and use AI large models to generate relevant, accurate and high-quality text; the disclosed embodiment can more effectively utilize AI models through the carefully designed Prompt to achieve accurate output of consumption types and consumption locations.

[0086] For example, a preferred prompt project is as follows:

[0087] 1) Design a consumption location prompt template:

[0088] You are an experienced travel guide with rich geographical knowledge and are well versed in itinerary planning and routes. Your task is to help users determine in which city each item in the given consumption name array occurs. If no city is included, answer "none";

[0089] Process the result into an array. The format of each object in the array is as follows:

[0090] 1. The value of the name field is of String type, and the value is the consumption name;

[0091] 2. The value of the city field is of String type, and the value is the city. Only the province and city need to be returned.

[0092] The user input is as follows:

[0093] ${consumptionData}

[0094] You just need to output in the specified format without any extra characters.

[0095] 2) Design a consumption type prompt word template:

[0096] You are a senior life manager with rich experience in consumption management. Your task is to help users identify which consumption category each item in the given consumption name array belongs to.

[0097] Now there are the following consumer categories:

[0098] ${consumptioneDefinition}

[0099] You need to process the result into an array, where each object in the array has the following format:

[0100] 1. The value of the name field is of String type, and the value is the consumption name;

[0101] 2. The value of the type field is of String type and must be one of the above consumption categories.

[0102] The user input is as follows:

[0103] ${consumptionData}

[0104] You just need to output in the specified format without any extra characters.

[0105] 3) Use the prompt word template to prompt the reasoning target of the big model and limit the reasoning scope, replace the consumption details data into the consumptionData variable of the consumption city prompt word template, and form the final question submitted to the big model. Use the basic common sense cognitive ability of the big model, and after the big model reasoning, return the consumption city corresponding to the consumption details data.

[0106] 4) Input the consumption location into the vector database, retrieve the definition of the consumption category from it, replace the consumption definition data into the consumptioneDefinition variable of the consumption type prompt word template, replace the consumption detail data into the consumptionData variable of the consumption type prompt word template, and input it into the big model. The big model infers the consumption type under the inference target of the prompt word.

[0107] 5) Collect the consumption type and consumption location data inferred by the large model and store them in a relational database.

[0108] The definition of consumption categories can be set according to actual conditions, such as catering, travel, housing, etc.; survival categories, development categories, and enjoyment categories; or necessities and non-necessities. In a preferred embodiment, the definition of consumption categories can be defined based on the retrieval of high-consumption and non-productive life-essential consumption in the consumption location, and high-consumption and non-productive life-essential consumption can be directly identified in the consumption type.

[0109] Furthermore, the method further comprises:

[0110] Develop prompt projects for obtaining property clues;

[0111] When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person to be executed.

[0112] In the same way as above, a prompt project can be formulated for property clues. Under the inference target of the prompt word, the big model can be used to infer property clues through consumption details. The property clue data is matched with the consumption type and consumption location and stored in a relational database.

[0113] A prompt project is formulated to obtain consumption type, consumption location, and property clues, and the prompt project is understood based on the big model to output the results. Compared with the current keyword matching method through mapping relationship tables, there is no need for basic data maintenance, and it also avoids the problem of inaccurate matching due to incomplete / wrong maintenance. AI technology can be better utilized to improve work efficiency and effectiveness.

[0114] Furthermore,

[0115] The output consumption results are matched with the target location and legal and regulatory information of the person subject to execution, and the high-consumption behaviors and consumption behaviors that are not necessary for production and life of the person subject to execution are determined, including:

[0116] Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination;

[0117] For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base;

[0118] The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location;

[0119] The report on the ability of the person subject to execution to generate the report based on the person subject to execution’s high-consumption behavior and non-productive and living consumption behavior includes:

[0120] Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

[0121] According to the output consumption results, the target place of the person to be executed and the legal and regulatory information are matched, the results are generated, and the report is assembled, including:

[0122] 1) Search the legal and regulatory knowledge base for the minimum living standard corresponding to the laws and regulations of the target place (place of household registration or residence) of the person subject to enforcement, and find out the records of consumption details that exceed the minimum living standard of that place. If the behavior of eating and staying in the place obviously exceeds the local consumption level, it will be judged as a high consumption record. At the same time, record the law and regulation as the basis for the report judgment.

[0123] 2) Take the consumption data processed by the large model and determine the details of consumption behaviors that are not necessary for production and life based on the consumption type. For example, the consumption type is taking airplanes, high-speed trains, etc.; buying expensive jewelry and jewelry, high-end watches, cosmetics and other luxury goods or buying cigarettes, alcohol, etc., buying non-business-essential vehicles, traveling, vacationing and other consumption behaviors; and extract the laws and regulations related to restricting the high consumption of the person subject to execution from the legal and regulatory knowledge base.

[0124] 3) Take the consumption data processed by the large model and accumulate the number of consumption times by consumption location.

[0125] 4) Take the consumption data processed by the large model, classify and summarize the property clues, and summarize the information such as the property location and vehicle name.

[0126] 5) Assemble the results of the above steps and generate a report on the performance ability of the person subject to execution, including the chain of evidence (details of abnormal consumption), basis for judgment (original text of regulations), etc.

[0127] According to the laws and regulations of the target place of the person to be executed in the individual case, high-consumption data is flexibly screened out, and the big model generates report conclusions based on the judgment results of high-consumption and non-essential consumption. Improve the efficiency and quality of report generation.

[0128] Furthermore, the method further comprises:

[0129] According to actual conditions, adjust the report data statistical time period and adapt local laws and regulations information, and regenerate the report.

[0130] For the consumption behavior of the person subject to enforcement who consumes a large amount of goods at places other than the target place, the report data statistical time period and local laws and regulations can be adjusted according to the actual situation, and the report can be regenerated, including:

[0131] 1) According to the time period when the case occurred, it supports adjusting the time range of consumption data retrieval to more accurately determine the consumption records during the period when the case occurred and regenerate the report.

[0132] 2) In case the high-frequency consumption locations of the person subject to execution are inconsistent with the target locations, the minimum living standard determined in the report can be adjusted to the local minimum living standard in the high-frequency consumption locations, and the report can be regenerated.

[0133] According to the frequent consumption locations of the person subject to enforcement, the local laws and regulations applicable to the report can be flexibly adjusted, the judgment basis of the report can be updated, and the report can be regenerated. This makes the report more professional and targeted. Achieving the goal of generating a more intelligent and accurate report on the ability to perform individual performance is of great significance to improving the efficiency of legal work.

[0134] The disclosed embodiment uses NLP to summarize and refine local laws and regulations, and obtain the basis for local judgments on minimum consumption. A prompt project is formulated to obtain consumption types, consumption locations, and property clues, and the prompt project is understood based on the big model to output the results. Compared with the current keyword matching method through mapping relationship tables, there is no need for basic data maintenance, and it also avoids the problem of inaccurate matching due to incomplete / wrong maintenance; when the report conclusion is generated, according to the laws and regulations of the target place of the person to be executed in each case, high-consumption data is individually screened out, and the report conclusion is provided by the big model to achieve "one case, one discussion". At the same time, it supports judges to flexibly adjust the local laws and regulations adapted to the report according to the high-frequency consumption locations of the person to be executed, so as to update the report results.

[0135] The second embodiment of the present disclosure further provides a method for generating a report on the performance ability of the person subject to execution based on a large model, including:

[0136] 1. Build a knowledge base of local laws and regulations

[0137] The process of building a legal and regulatory knowledge base is as follows: Figure 2 As shown, legal and regulatory documents related to minimum living security from all over the country / the Supreme Court are input, and NLP is used to understand and refine the language of the minimum living security standards in the laws and regulations. The key information of the minimum living security standards for each year is extracted by location, and the refined location text, year, and minimum living security standard numbers are saved in the legal and regulatory knowledge base for subsequent use.

[0138] 2. Upload mobile payment details and clean the data

[0139] Enter the mobile payment details data, perform data cleaning, and remove the records marked with "payment failed, refund successful" in the mobile payment details, as well as their associated income / expenditure records.

[0140] 3. Upload case judgment documents and extract key case information

[0141] Figure 3 Construct a flow chart for the case information database, such as Figure 3 As shown, the process includes inputting the case judgment document, using NLP to understand and segment the case information, extracting key information such as the file number, case subject, name of the person to be executed, and place of residence, and saving it in the case information database for subsequent use.

[0142] 4. Input consumption details, and the big model will understand the consumption type, consumption location, and property clues, and output the results

[0143] According to the business scenario of the performance report of the person subject to execution, prompt projects are formulated for obtaining consumption types, consumption locations, and property clues. According to the output results, repeated debugging is carried out to finally form a prompt project with clear background information and output instructions. After entering the consumption details, the prompt project is understood based on the big model and the fixed format results are output.

[0144] 1) Design a consumption location prompt template:

[0145] You are an experienced travel guide with rich geographical knowledge and are well versed in itinerary planning and routes. Your task is to help users determine in which city each item in the given consumption name array occurs. If no city is included, answer "none";

[0146] Process the result into an array. The format of each object in the array is as follows:

[0147] 1. The value of the name field is of String type, and the value is the consumption name;

[0148] 2. The value of the city field is of String type, and the value is the city. Only the province and city need to be returned.

[0149] The user input is as follows:

[0150] ${consumptionData}

[0151] You just need to output in the specified format without any extra characters.

[0152] 2) Design a consumption type prompt word template:

[0153] You are a senior life manager with rich experience in consumption management. Your task is to help users identify which consumption category each item in the given consumption name array belongs to.

[0154] Now there are the following consumer categories:

[0155] ${consumptioneDefinition}

[0156] You need to process the result into an array, where each object in the array has the following format:

[0157] 1. The name field value is of String type, and the value is the consumption name;

[0158] 2. The value of the type field is of String type and must be one of the above consumption categories.

[0159] The user input is as follows:

[0160] ${consumptionData}

[0161] You just need to output in the specified format without any extra characters.

[0162] 3) Use the prompt word template to prompt the reasoning target of the big model and limit the reasoning scope, replace the consumption details data into the consumptionData variable of the consumption city prompt word template, and form the final question submitted to the big model. Use the basic common sense cognitive ability of the big model, and after the big model reasoning, return the consumption city corresponding to the consumption details data.

[0163] 4) Input the consumption location into the vector database, retrieve the consumption definitions of high consumption and non-productive life in the consumption location, replace the consumption definition data into the consumptioneDefinition variable of the consumption type prompt word template, replace the consumption detail data into the consumptionData variable of the consumption type prompt word template, and input it into the big model. The big model infers the consumption type and property clues under the inference target of the prompt word.

[0164] 5) Collect consumption type, consumption location, and property clue data inferred by the large model and store them in a relational database.

[0165] 5. Match the domicile of the person subject to execution and the legal and regulatory information, generate results, and assemble reports; the report generation flow chart is as follows: Figure 4 ;

[0166] 1) Search the legal and regulatory knowledge base for the minimum living standard corresponding to the laws and regulations of the place of residence of the person subject to execution. Any record of consumption details that exceeds the minimum living standard of the place is considered a high-consumption record. At the same time, record the law and regulation as the basis for the report.

[0167] 2) Take the consumption data processed by the big model and determine the details of consumption behaviors that are not necessary for production and life based on the consumption type. And formulate a prompt project to extract the laws and regulations related to restricting the high consumption of the person subject to execution from the legal and regulatory knowledge base.

[0168] 3) Take the consumption data processed by the large model and accumulate the number of consumption times by consumption location.

[0169] 4) Take the consumption data processed by the large model, classify and summarize the property clues, and summarize the information such as the property location and vehicle name.

[0170] 5) Assemble the results of the above steps and generate a report on the performance ability of the person subject to execution, including the chain of evidence (details of abnormal consumption), basis for judgment (original text of regulations), etc.

[0171] 6. Support adjusting the statistical time period of report data and adapting local laws and regulations according to actual conditions, and regenerating reports

[0172] 1) According to the time period when the case occurred, it supports adjusting the time range of consumption data retrieval to more accurately determine the consumption records during the period when the case occurred and regenerate the report.

[0173] 2) In case the debtor’s high-frequency consumption locations are inconsistent with their registered residence, the minimum living standard determined in the report can be adjusted to the local minimum living standard in the high-frequency consumption locations, and the report can be regenerated.

[0174] The disclosed embodiment adopts an innovative information acquisition mode: a prompt project is formulated for obtaining consumption types, consumption locations, and property clues, and the prompt project is understood based on the big model to output the results. Compared with the current keyword matching method through mapping relationship tables, there is no need for basic data maintenance, which avoids the problem of inaccurate matching due to incomplete / wrong maintenance. According to the laws and regulations of the target location of the person to be executed in each case, high-consumption data is individually screened out, and the big model provides report conclusions to achieve "one case, one discussion". At the same time, it supports judges to flexibly adjust the local laws and regulations adapted to the report according to the high-frequency consumption locations of the person to be executed, update the report judgment basis, and regenerate the report. Achieving the goal of generating personal performance ability reports more intelligently and accurately is of great significance to improving the efficiency of legal work.

[0175] The third embodiment of the present disclosure also provides a system for generating a report on the performance ability of the person subject to execution based on a large model, such as Figure 5 As shown, the system comprises:

[0176] An upload module 11, which is configured to upload payment details of the person subject to execution and perform data cleaning;

[0177] An extraction module 12, which is configured to extract key case information, including file number, case subject, name of the person to be executed, and target location, based on the case judgment document;

[0178] The data processing module 13 is configured to input the payment details after data cleaning, and the big model understands the consumption type and consumption location, and outputs the consumption results;

[0179] The determination module 14 is configured to match the output consumption results with the destination of the person subject to execution and the legal and regulatory information, and determine the high-consumption behavior and the consumption behavior that is not necessary for production and life of the person subject to execution;

[0180] The generation module 15 is configured to generate a report on the performance ability of the person subject to execution based on the person's high consumption behaviors and consumption behaviors that are not necessary for production and life.

[0181] Furthermore, the system also includes:

[0182] The extraction module 16 is configured to input documents related to the minimum living security, use NLP to extract the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the extracted location text, year, and minimum living security standard numbers into the knowledge base.

[0183] Furthermore, the data processing module 13 is specifically configured as follows:

[0184] According to the business scenario of the performance ability report of the person subject to execution, a prompt project is formulated for obtaining consumption type and consumption location respectively;

[0185] Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions;

[0186] After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

[0187] Furthermore, the data processing module 13 is also configured to:

[0188] Develop prompt projects for obtaining property clues;

[0189] When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person subject to execution.

[0190] Furthermore,

[0191] The determination module 14 is specifically configured as follows:

[0192] Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination;

[0193] For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base;

[0194] The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location;

[0195] The generating module 15 is specifically configured as follows:

[0196] Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

[0197] Furthermore, the system also includes an updating module 17;

[0198] The updating module 17 is configured to adjust the statistical time period of the report data and adapt the local laws and regulations according to the actual situation, so that the generating module regenerates the report.

[0199] The system for generating a report on the ability of the person to be executed based on a big model in the embodiment of the present disclosure is used to implement the method for generating a report on the ability of the person to be executed based on a big model in method embodiment 1 and embodiment 2, so the description is relatively simple. For details, please refer to the relevant description in the previous method embodiment, which will not be repeated here.

[0200] In addition, if Figure 6 As shown, the fourth embodiment of the present disclosure further provides an electronic device, including a memory 100 and a processor 200, wherein the memory 100 stores a computer program, and when the processor 200 runs the computer program stored in the memory 100, the processor 200 executes the above-mentioned various possible methods.

[0201] The memory 100 is connected to the processor 200. The memory 100 may be a flash memory, a read-only memory or other memory, and the processor 200 may be a central processing unit or a single-chip microcomputer.

[0202] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the above-mentioned various possible methods.

[0203] 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 technology, CD-ROM (Compact Disc Read-Only Memory), Digital Video Disc (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0204] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for generating a report on the performance ability of a person subject to execution based on a large model, characterized in that: The method comprises: Upload the payment details of the person subject to execution and clean the data; Extract key case information based on the case judgment documents, including the case file number, case subject, name of the person to be executed, and target location; Input payment details after data cleaning, and the big model will understand the consumption type and location, and output the consumption results; Match the output consumption results with the target location and legal and regulatory information of the person subject to execution, and determine the person's high-consumption behavior and consumption behavior that is not necessary for production and life; A report on the debtor's ability to perform is generated based on the debtor's high consumption behavior and consumption behavior that is not necessary for production and life.

2. The method according to claim 1, characterized in that The method further comprises: Input documents related to minimum living security, use natural language processing (NLP) to extract the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the extracted location text, year, and minimum living security standard numbers into the knowledge base.

3. The method according to claim 2, characterized in that The payment details data after the input data cleaning is understood by the big model based on the consumption type and consumption location, and the consumption results are output, including: According to the business scenario of the performance ability report of the person subject to execution, an AI large model prompt word project is developed for obtaining consumption types and consumption locations; Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions; After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

4. The method according to claim 3, characterized in that: The method further comprises: Develop prompt projects for obtaining property clues; When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person to be executed.

5. The method according to claim 4, characterized in that The output consumption results are matched with the target location and legal and regulatory information of the person subject to execution, and the high-consumption behaviors and consumption behaviors that are not necessary for production and life of the person subject to execution are determined, including: Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination; For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base; The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location; The report on the ability of the person subject to execution to generate the report based on the person subject to execution’s high-consumption behavior and non-productive and living consumption behavior includes: Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

6. The method according to claim 1, characterized in that The method further comprises: According to actual conditions, adjust the report data statistical time period and adapt local laws and regulations information, and regenerate the report.

7. A system for generating a report on the performance ability of the person subject to execution based on a large model, characterized in that: The system comprises: The upload module is configured to upload payment details of the person subject to execution and perform data cleansing; An extraction module is configured to extract key case information, including file number, case subject, name of the person to be executed, and target location, based on the case judgment document; The data processing module is set to input the payment details after data cleaning, and the big model understands the consumption type and consumption location, and outputs the consumption results; A determination module is configured to match the output consumption results with the destination of the person subject to execution and the legal and regulatory information, and determine the high-consumption behavior and consumption behavior that is not necessary for production and life of the person subject to execution; A generation module is configured to generate a report on the performance ability of the person subject to execution based on the person's high consumption behavior and consumption behavior that is not necessary for production and life.

8. The system according to claim 7, characterized in that The system further comprises: The extraction module is set to input documents related to the minimum living security, use NLP to extract the provisions of the minimum living security standards in the documents, extract the key information of the minimum living security standards for each year by location, and save the extracted location text, year, and minimum living security standard numbers into the knowledge base.

9. The system according to claim 8, characterized in that The data processing module is specifically configured as follows: According to the business scenario of the performance ability report of the person subject to execution, a prompt project is formulated for obtaining consumption type and consumption location respectively; Debug the prompt project based on the output results, and eventually form a prompt project with clear background information and output instructions; After entering the payment details, the consumption results in a fixed format are output based on the big model's understanding of prompt engineering.

10. The system according to claim 9, characterized in that The data processing module is also configured to: Develop prompt projects for obtaining property clues; When outputting the consumption results, the property clue results are output, and the property clues are classified and summarized according to the output property clue results to summarize the real estate information and vehicle information of the person to be executed.

11. The system according to claim 10, characterized in that The determination module is specifically configured as follows: Search the knowledge base for the corresponding minimum living standard in the laws and regulations of the target place of the person subject to execution, and determine the records of consumption results that occur in the place and exceed the minimum living standard of the place as high consumption behavior, and record the laws and regulations of the target place as the basis for report determination; For the consumption results output by the big model, the consumption behaviors that are not necessary for production and life are determined based on the consumption types, and the laws and regulations related to restricting the consumption that is not necessary for production and life of the person subject to execution are extracted from the knowledge base; The number of high-consumption behaviors and the number of consumption behaviors that are not necessary for production and life are accumulated by consumption location; The generation module is specifically configured as follows: Generate a report on the debtor's ability to perform, including a chain of evidence and corresponding basis for judgment. The chain of evidence includes the debtor's high-consumption behavior, consumption behavior that is not necessary for production and life, and the debtor's real estate information and vehicle information.

12. The system according to claim 7, characterized in that The system also includes an update module; The updating module is configured to adjust the statistical time period of the report data and adapt the local laws and regulations information according to the actual situation, so that the generating module regenerates the report.

13. An electronic device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the method for generating a report on the performance ability of the person subject to execution based on a large model as described in any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for generating a report on the performance ability of the person subject to execution based on a large model as described in any one of claims 1 to 6.

Images