A method and system for generating public health supervision and law enforcement documents by artificial intelligence (AI) agent

Abstract

The present invention discloses a method and system for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent, and relates to the technical field of public health supervision and law enforcement data processing. The system comprises: a data acquisition module, a data calculation module, a model processing module and an artificial intelligence (AI) agent module. The algorithm comprises: real-time acquisition of data information in the process of health supervision and law enforcement events, receiving a health supervision and law enforcement document generation instruction, calling a supervision and law enforcement document model, generating an internal supervision and law enforcement document, calling an external tool, generating an external supervision and law enforcement document, jointly analyzing the internal supervision and law enforcement document and the external supervision and law enforcement document, optimizing the supervision and law enforcement document model and generating the optimal supervision and law enforcement document, so as to solve the problems that it is easy to miss or misremember law enforcement event elements when manually writing law enforcement documents, the system preset supervision and law enforcement document model is not optimized in time, the law enforcement document generation efficiency is low, the content is inaccurate, and the terminology is not standardized.

Description

Technical Field

[0001] The present invention belongs to the technical field of public health supervision and law enforcement data processing, and specifically relates to a method and system for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent. Background Art

[0002] Artificial intelligence AI agent is a higher-level technology of artificial intelligence. Its essential characteristic that distinguishes it from traditional artificial intelligence is that artificial intelligence AI agent can make autonomous decisions based on the information obtained, and complete core functions such as information processing, reasoning, and decision-making. At the same time, artificial intelligence AI agent can use tools autonomously.

[0003] In the process of health supervision and law enforcement, the writing of health supervision and law enforcement documents is usually done by law enforcement personnel after the law enforcement incident is completed or is about to be completed. Since law enforcement incidents are usually complex and sudden, it is easy for law enforcement personnel to make omissions or mistakes when writing law enforcement documents after the law enforcement incident is completed or is about to be completed. At the same time, due to the differences in the professional levels of law enforcement personnel and the differences in document writing abilities, it is difficult to ensure the accuracy of the content and the standardization of the terms in manually written supervision and law enforcement documents. Although the existing supervision and law enforcement auxiliary systems also have the function of automatically generating supervision and law enforcement documents, this function is mostly achieved by automatically replacing the keywords or key sentences of the corresponding positions in the system's preset document templates with the keywords or key sentences of the law enforcement incidents manually input by the law enforcement personnel. The law enforcement documents written in this way have rigid and stiff sentences, serious formalization of the entire text, and are unable to identify wrong words and incorrect sentences in the documents, which seriously affects the law enforcement image of the law enforcement personnel. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent, which utilizes the artificial intelligence (AI) agent to autonomously use tools to complete information processing, reasoning, and decision-making functions, so as to solve the problems of omissions, errors, inaccurate content, and non-standard terms in manually writing law enforcement documents in the process of health supervision and law enforcement, as well as the problem that the supervision and law enforcement documents automatically generated by the supervision and law enforcement auxiliary system have rigid and stiff sentences, formalized formatting of the entire text, and cannot recognize typos and incorrect sentences in the text.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent, comprising the following steps:

[0007] S100, collecting data information during the health supervision law enforcement event in real time, and storing the data information according to the law enforcement event category;

[0008] S101. Check whether the instruction to generate health supervision and law enforcement documents has been issued;

[0009] S102, receiving a health supervision and law enforcement document generation instruction, calling a supervision and law enforcement document model based on the collected health supervision and law enforcement event data information, and generating an internal supervision and law enforcement document;

[0010] S103. Based on the collected health supervision and law enforcement event data information, call external tools to generate external supervision and law enforcement documents;

[0011] S104. Conduct joint analysis on internal supervision and law enforcement documents and external supervision and law enforcement documents to generate the optimal supervision and law enforcement documents;

[0012] S105. Optimize and store supervision and law enforcement document models;

[0013] Among them, step S102 receives the instruction to generate the health supervision and law enforcement document, calls the supervision and law enforcement document model according to the collected health supervision and law enforcement event data information, and generates the process of internal supervision and law enforcement documents, including the following steps:

[0014] S201. Preset supervisory and law enforcement document style models, including commonly used supervisory and law enforcement document style models and special supervisory and law enforcement document style models;

[0015] S202, reading the real-time data information of the supervision and law enforcement of the law enforcement case, and preprocessing the data information to form a supervision and law enforcement data information set, calculating the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determining the optimal similarity text set ∑S(i);

[0016] S203, according to the optimal similarity text set ∑S(i), retrieve the corresponding supervision and law enforcement text segment in the database to generate a first draft of the supervision and law enforcement text;

[0017] S204. Sort the text paragraphs in the draft of the supervision and law enforcement text by relevance, and calculate the law enforcement document text legitimacy coefficient for the text after relevance sorting;

[0018] S205. Determine the legality of law enforcement documents and generate internal supervision law enforcement documents;

[0019] Wherein, in step S202, the calculation formula of the text similarity coefficient S(i, j) is:

[0020]

[0021] Where, S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, r is a hyperparameter, and the value range of r is 0.1≤r≤0.6;

[0022] The calculation formula of the text legitimacy coefficient Hlxse in step S204 is:

[0023]

[0024] In the formula, Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, and k is the adjustment factor.

[0025] Preferably, step S202 reads the real-time data information of the supervision and law enforcement of the law enforcement case, pre-processes the data information, forms a supervision and law enforcement data information set, calculates the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determines the process of the optimal similarity text set ∑S(i), including the following steps:

[0026] S21. Read the supervision and law enforcement data information of law enforcement cases in real time;

[0027] S22. Mark the supervision and law enforcement data information of law enforcement cases. The marking categories include: party information, law enforcement agency information, case information, case information, violation information, penalty opinion information and dispute points information;

[0028] S23, classify and integrate the marked supervision and law enforcement data information, remove useless data information, and merge duplicate data information;

[0029] S24, extract the integrated data text segments, calculate the text similarity coefficient S(i, j) of each text segment respectively, and generate the optimal similarity text set ∑S(i).

[0030] Preferably, the process of generating the law enforcement document legitimacy coefficient in step S204 includes the following steps:

[0031] S41, presetting a data text relevance coefficient model and a data text legitimacy coefficient model;

[0032] S42. Calculate the correlation coefficient of each text segment in the draft of the supervision and law enforcement document;

[0033] S43, using a preset data text correlation coefficient model, sorting and condensing the correlation coefficients of the text paragraphs in the draft of the supervision and law enforcement document;

[0034] S44. Calculate the text legitimacy coefficient of the first draft of the supervision and law enforcement document through the data text legitimacy coefficient model;

[0035] S45. Determine the legality coefficient of the text of the first draft of the supervisory and law enforcement document and determine the internal supervisory and law enforcement document.

[0036] Preferably, in step S42, in calculating the correlation coefficient of each text segment in the draft of the supervision and law enforcement document, the correlation coefficient Glxse(w, i) is calculated by the formula:

[0037]

[0038] Where Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1, k2 are constants, and k1 = 0.2, k2 = 0.5;

[0039] Preferably, step S104 performs a joint analysis on the internal supervision and law enforcement documents and the external supervision and law enforcement documents, and the process of generating the optimal supervision and law enforcement documents includes the following steps:

[0040] S401. Retrieve internal supervision and law enforcement documents and external supervision and law enforcement documents respectively;

[0041] S402. Conduct joint analysis on internal and external supervisory and law enforcement documents to optimize the wording and layout of the text;

[0042] S403. Generate optimal supervisory and law enforcement documents.

[0043] A system for generating public health supervision and law enforcement documents using an artificial intelligence (AI) agent, comprising:

[0044] The data collection module is used to collect the law enforcement data information of health supervision and law enforcement events, and pre-process the data information to form a supervision and law enforcement data information set, and send the supervision and law enforcement data information set to the data processing module;

[0045] The data processing module is used to mark, classify, and integrate the law enforcement data information of health supervision law enforcement events, calculate the text similarity coefficient S(i, j) of each text segment of the law enforcement data information, generate the optimal similarity text set ∑S(i), and send the optimal similarity text set ∑S(i) to the model operation module, wherein the calculation formula of the text similarity coefficient S(i, j) is:

[0046]

[0047] Where, S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, r is a hyperparameter, and the value range of r is 0.1≤r≤0.6;

[0048] The model operation module is used to store the data text correlation coefficient model and the data text legitimacy coefficient model, calculate the correlation coefficient Glxse(w, i) of each text segment in the draft of the supervision and law enforcement document and the text legitimacy coefficient Hlxse of the draft of the supervision and law enforcement document through the data text correlation coefficient model and the data text legitimacy coefficient model, judge the legitimacy of the supervision and law enforcement document, form an internal supervision and law enforcement document, and send the internal supervision and law enforcement document to the database, wherein, in step S204, the correlation coefficient Glxse(w, i) of the text segment is used in the correlation sorting of each text segment in the draft of the supervision and law enforcement text, and the calculation formula is:

[0049]

[0050] Where Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1, k2 are constants, and k1 = 0.2, k2 = 0.5;

[0051] The calculation formula of the text legitimacy coefficient Hlxse in step S204 is:

[0052]

[0053] Where Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, k is the adjustment factor;

[0054] The artificial intelligence AI agent module is used to execute the artificial intelligence AI agent program, including calling external tools based on the collected health supervision and law enforcement event data information, generating external supervision and law enforcement documents, conducting joint analysis of internal supervision and law enforcement documents and external supervision and law enforcement documents, generating optimal supervision and law enforcement documents, and optimizing and storing supervision and law enforcement document models.

[0055] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0056] 1. The system collects data information of law enforcement events in real time, and completes the preprocessing of data information and the calculation of similarity coefficients of data information text segments, which can avoid omissions and missing of law enforcement data information and ensure that all data information of law enforcement events is recorded completely and effectively.

[0057] 2. By further processing the similarity coefficients of data information text segments, the optimal similarity text set ∑S(i) of data information text segments is generated, and the corresponding supervision and law enforcement text segments in the database are retrieved. The system can automatically generate a draft of the supervision and law enforcement document, and as the law enforcement events progress, the law enforcement data information is constantly improved, and the draft of the supervision and law enforcement document is also synchronously updated and improved. Compared with law enforcement documents written by law enforcement personnel, the accuracy and efficiency are greatly improved.

[0058] 3. The present invention presets a data-text correlation coefficient model and a data-text legitimacy coefficient model. By calculating the correlation coefficient of each text segment in the draft of the supervision and law enforcement document and the text legitimacy coefficient of the draft of the supervision and law enforcement document, the order of each segment in the law enforcement document, the standardization and readability of the writing can be further optimized, the formatting of the law enforcement document can be avoided, and at the same time, wrong words and incorrect sentences in the document can be eliminated.

[0059] 4. Taking advantage of the ability of artificial intelligence (AI) agents to autonomously use external tools, use Internet resources to generate external supervisory and law enforcement documents, and combine them with internal supervisory and law enforcement documents generated by preset law enforcement document models to improve the generation methods and efficiency of supervisory and law enforcement documents.

[0060] 5. Artificial intelligence AI agents can make autonomous decisions based on the information obtained, complete the characteristics of core functions such as information processing, reasoning, and decision-making, optimize the preset supervision and law enforcement models in the system, and improve the efficiency and standardization of the generation of supervision and law enforcement documents. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A flowchart of a method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent of the present invention is shown;

[0063] Figure 2A system module diagram of a method for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent of the present invention is shown. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Example 1

[0066] See also Figure 1 As shown, this embodiment provides a method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent, comprising the following steps:

[0067] S100, collecting data information during the health supervision law enforcement event in real time, and storing the data information according to the law enforcement event category;

[0068] S101. Check whether the instruction to generate health supervision and law enforcement documents has been issued;

[0069] S102, receiving a health supervision and law enforcement document generation instruction, calling a supervision and law enforcement document model based on the collected health supervision and law enforcement event data information, and generating an internal supervision and law enforcement document;

[0070] S103. Based on the collected health supervision and law enforcement event data information, call external tools to generate external supervision and law enforcement documents;

[0071] S104. Conduct joint analysis on internal supervision and law enforcement documents and external supervision and law enforcement documents to generate the optimal supervision and law enforcement documents;

[0072] S105. Optimize and store the supervision and law enforcement document model.

[0073] Among them, step S102 receives the instruction to generate the health supervision and law enforcement document, calls the supervision and law enforcement document model according to the collected health supervision and law enforcement event data information, and generates the process of internal supervision and law enforcement documents, including the following steps:

[0074] S201. Preset supervisory and law enforcement document style models, including commonly used supervisory and law enforcement document style models and special supervisory and law enforcement document style models;

[0075] S202, reading the real-time data information of the supervision and law enforcement of the law enforcement case, and preprocessing the data information to form a supervision and law enforcement data information set, calculating the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determining the optimal similarity text set ∑S(i);

[0076] S203, according to the optimal similarity text set ∑S(i), retrieve the corresponding supervision and law enforcement text segment in the database to generate a first draft of the supervision and law enforcement text;

[0077] S204. Sort the text paragraphs in the draft of the supervision and law enforcement text by relevance, and calculate the law enforcement document legitimacy coefficient for the entire text after relevance sorting;

[0078] S205. Determine the legality of law enforcement documents and generate internal supervision law enforcement documents.

[0079] Wherein, in step S202, the calculation formula of the text similarity coefficient S(i, j) is:

[0080]

[0081] Where S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, and r is a hyperparameter whose value range is 0.1≤r≤0.6.

[0082] It should be noted that the purpose of calculating the text similarity coefficient of the text segment is to find the most similar standard text in the database for the data information of the law enforcement event collected by the system, that is, the candidate text segment corresponding to the maximum similarity coefficient value is the candidate text segment of the data information of the law enforcement event;

[0083] The calculation formula of the text legitimacy coefficient Hlxse in step S204 is:

[0084]

[0085] In the formula, Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, and k is the adjustment factor.

[0086] It should be noted that the calculation of the text legitimacy coefficient is a comprehensive review of the wording standardization of the entire text of the supervision and law enforcement document, the degree of sentence sequence and the completeness of the law enforcement event elements. The higher the text legitimacy coefficient value, the more complete the supervision and law enforcement document is;

[0087] Step S202 reads the real-time data information of the supervision and law enforcement of the law enforcement case, pre-processes the data information, forms a supervision and law enforcement data information set, calculates the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determines the process of the optimal similarity text set ∑S(i), including the following steps:

[0088] S21. Read the supervision and law enforcement data information of law enforcement cases in real time;

[0089] S22. Mark the supervision and law enforcement data information of law enforcement cases. The marking categories include: party information, law enforcement agency information, case information, case information, violation information, penalty opinion information and dispute points information;

[0090] S23, classify and integrate the marked supervision and law enforcement data information, remove useless data information, and merge duplicate data information;

[0091] The specific classification and integration process is as follows: collect the supervision and law enforcement information with the same tag category, and then screen the information in the collection, remove useless data information, and merge data information with duplicate content;

[0092] S24, extract the integrated data text segments, calculate the text similarity coefficient S(i, j) of each text segment respectively, and generate the optimal similarity text set ∑S(i).

[0093] The process of generating the law enforcement document legitimacy coefficient in step S204 includes the following steps:

[0094] S41, presetting a data text relevance coefficient model and a data text legitimacy coefficient model;

[0095] S42. Calculate the correlation coefficient of each text segment in the draft of the supervision and law enforcement document;

[0096] S43. Using a preset data text correlation coefficient model, sort and condense the text paragraphs in the draft of the supervision and law enforcement document according to the correlation coefficient, so that the text paragraphs in the draft of the supervision and law enforcement document and the paragraphs before and after it are re-sorted from large to small according to the correlation coefficient value, and delete and replace redundant text information according to language specifications;

[0097] S44. Calculate the text legitimacy coefficient of the first draft of the supervision and law enforcement document through the data text legitimacy coefficient model;

[0098] S45. Determine the legality coefficient of the text of the first draft of the supervisory and law enforcement document and determine the internal supervisory and law enforcement document.

[0099] Among them, in step S42, in calculating the correlation coefficient of each text segment in the draft of the supervision and law enforcement document, the correlation coefficient Glxse(w, i) is calculated by the formula:

[0100]

[0101] Where Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1, k2 are constants, and K1 = 0.2, k2 = 0.5;

[0102] It should be noted that the greater the correlation coefficient of text segments, the greater the correlation between the two text segments, that is, the greater the possibility that the two text segments tend to have the same semantics, which is reflected in the text form as the smaller the distance between the two text segments. By comparing the correlation coefficient values ​​between text segments, the natural sorting of text segments can be automatically completed;

[0103] Step S104 performs a joint analysis on the internal supervision and enforcement documents and the external supervision and enforcement documents to generate the optimal supervision and enforcement documents, including the following steps:

[0104] S401. Retrieve internal supervision and law enforcement documents and external supervision and law enforcement documents respectively;

[0105] S402. Conduct joint analysis on internal and external supervisory and law enforcement documents to optimize the wording and layout of the text;

[0106] S403. Generate optimal supervisory and law enforcement documents.

[0107] Example 2

[0108] See also Figure 2 As shown, this embodiment provides a system for generating public health supervision and law enforcement documents by an artificial intelligence AI agent, including:

[0109] The data collection module is used to collect the law enforcement data information of health supervision and law enforcement events, and pre-process the data information to form a supervision and law enforcement data information set, and send the supervision and law enforcement data information set to the data processing module;

[0110] The data processing module is used to mark, classify, and integrate the law enforcement data information of health supervision law enforcement events, calculate the text similarity coefficient S(i, j) of each text segment of the law enforcement data information, generate the optimal similarity text set ∑S(i), and send the optimal similarity text set ∑S(i) to the model operation module, wherein the calculation formula of the text similarity coefficient S(i, j) is:

[0111]

[0112] Where, S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, r is a hyperparameter, and the value range of r is 0.1≤r≤0.6;

[0113] The model operation module is used to store the data text correlation coefficient model and the data text legitimacy coefficient model, calculate the correlation coefficient Glxse(w, i) of each text segment in the draft of the supervision and law enforcement document and the text legitimacy coefficient Hlxse of the draft of the supervision and law enforcement document through the data text correlation coefficient model and the data text legitimacy coefficient model, judge the legitimacy of the supervision and law enforcement document, form an internal supervision and law enforcement document, and send the internal supervision and law enforcement document to the database, wherein, in step S204, the correlation coefficient Glxse(w, i) of the text segment is used in the correlation sorting of each text segment in the draft of the supervision and law enforcement text, and the calculation formula is:

[0114]

[0115] Where Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1, k2 are constants, and k1 = 0.2, k2 = 0.5;

[0116] The calculation formula of the text legitimacy coefficient Hlxse in step S204 is:

[0117]

[0118] Where Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, k is the adjustment factor;

[0119] The artificial intelligence AI agent module is used to execute the artificial intelligence AI agent program, including calling external tools based on the collected health supervision and law enforcement event data information, generating external supervision and law enforcement documents, conducting joint analysis of internal supervision and law enforcement documents and external supervision and law enforcement documents, generating optimal supervision and law enforcement documents, and optimizing and storing supervision and law enforcement document models.

[0120] The formulas mentioned in this application are obtained by collecting a large amount of data, removing the dimensions, and performing software simulation to obtain a formula that is close to the actual value. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The thresholds and preset range values ​​mentioned in this application are obtained and selected by software communication technicians after collecting a large amount of data and performing software simulation.

[0121] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. The functional units in each embodiment of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0122] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for generating public health supervision and law enforcement documents by an artificial intelligence (AI) agent, characterized in that: The following steps are involved: S100, collecting data information during the health supervision law enforcement event in real time, and storing the data information according to the law enforcement event category; S101. Check whether the instruction to generate health supervision and law enforcement documents has been issued; S102, receiving a health supervision and law enforcement document generation instruction, calling a supervision and law enforcement document model based on the collected health supervision and law enforcement event data information, and generating an internal supervision and law enforcement document; S103. Based on the collected health supervision and law enforcement event data information, call external tools to generate external supervision and law enforcement documents; S104. Conduct joint analysis on internal supervision and law enforcement documents and external supervision and law enforcement documents to generate the optimal supervision and law enforcement documents; S105. Optimize and store supervision and law enforcement document models; Among them, step S102 receives the instruction to generate the health supervision and law enforcement document, calls the supervision and law enforcement document model according to the collected health supervision and law enforcement event data information, and generates the process of internal supervision and law enforcement documents, including the following steps: S201. Preset supervisory and law enforcement document style models, including commonly used supervisory and law enforcement document style models and special supervisory and law enforcement document style models; S202, reading the real-time data information of the supervision and law enforcement of the law enforcement case, and preprocessing the data information to form a supervision and law enforcement data information set, calculating the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determining the optimal similarity text set ∑S(i); S203, according to the optimal similarity text set ∑S(i), retrieve the corresponding supervision and law enforcement text segment in the database to generate a first draft of the supervision and law enforcement text; S204. Sort the text paragraphs in the draft of the supervision and law enforcement text by relevance, and calculate the law enforcement document text legitimacy coefficient for the text after relevance sorting; S205. Determine the legality of law enforcement documents and generate internal supervision law enforcement documents; Wherein, in step S202, the calculation formula of the text similarity coefficient S(i, j) is: Where, S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, r is a hyperparameter, and the value range of r is 0.1≤r≤0.6; The calculation formula of the text legitimacy coefficient Hlxse in step S204 is: In the formula, Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, and k is the adjustment factor.

2. According to the method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent according to claim 1, it is characterized in that: Step S202 reads the real-time data information of the supervision and law enforcement of the law enforcement case, pre-processes the data information, forms a supervision and law enforcement data information set, calculates the text similarity coefficient S(i, j) of each information data in the supervision and law enforcement data information set, and determines the process of the optimal similarity text set ∑S(i), including the following steps: S21. Read the supervision and law enforcement data information of law enforcement cases in real time; S22. Mark the supervision and law enforcement data information of law enforcement cases. The marking categories include: party information, law enforcement agency information, case information, case information, violation information, penalty opinion information and dispute points information; S23, classify and integrate the marked supervision and law enforcement data information, remove useless data information, and merge duplicate data information; S24, extract the integrated data text segments, calculate the text similarity coefficient S(i, j) of each text segment respectively, and generate the optimal similarity text set ∑S(i).

3. According to the method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent according to claim 1, it is characterized in that: In step S204, the process of sorting the relevance of each text segment in the draft of the supervision and law enforcement text and calculating the legality coefficient of the law enforcement document text for the entire text after the relevance sorting includes the following steps: S41, presetting a data text relevance coefficient model and a data text legitimacy coefficient model; S42. Calculate the correlation coefficient of each text segment in the draft of the supervision and law enforcement document; S43, using a preset data text correlation coefficient model, sorting and condensing the correlation coefficients of the text paragraphs in the draft of the supervision and law enforcement document; S44. Calculate the text legitimacy coefficient of the first draft of the supervisory and law enforcement document through the data text legitimacy coefficient model.

4. According to claim 3, a method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent is characterized in that: In step S42, in calculating the correlation coefficient of each text segment in the draft of the supervision and law enforcement document, the correlation coefficient of the text segment Glxse(w, i) is calculated by the formula: In the formula, Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1 and k2 are constants, and k1=0.2 and k2=0.

5.

5. According to the method for generating public health supervision and law enforcement documents by an artificial intelligence AI agent according to claim 1, it is characterized in that: Step S104 performs a joint analysis on the internal supervision and enforcement documents and the external supervision and enforcement documents to generate the optimal supervision and enforcement documents, including the following steps: S401. Retrieve internal supervision and law enforcement documents and external supervision and law enforcement documents respectively; S402. Conduct joint analysis on internal and external supervisory and law enforcement documents to optimize the wording and layout of the text; S403. Generate optimal supervisory and law enforcement documents.

6. A system for generating public health supervision and law enforcement documents using an artificial intelligence (AI) agent, characterized in that: include: The data collection module is used to collect the law enforcement data information of health supervision and law enforcement events, and pre-process the data information to form a supervision and law enforcement data information set, and send the supervision and law enforcement data information set to the data processing module; The data processing module is used to mark, classify, and integrate the law enforcement data information of health supervision law enforcement events, calculate the text similarity coefficient S(i, j) of each text segment of the law enforcement data information, generate the optimal similarity text set ∑S(i), and send the optimal similarity text set ∑S(i) to the model operation module, wherein the calculation formula of the text similarity coefficient S(i, j) is: Where, S(i, j) is the text similarity coefficient between text segment i and candidate text segment j; S1 is the number of bytes that are completely identical between text segment i and candidate text segment j; S2 is the number of bytes that are not completely identical between text segment i and candidate text segment j; pj is the total number of bytes of candidate text segment j, r is a hyperparameter, and the value range of r is 0.1≤r≤0.6; The model operation module is used to store the data text correlation coefficient model and the data text legitimacy coefficient model, calculate the correlation coefficient Glxse(w, i) of each text segment in the draft of the supervision and law enforcement document and the text legitimacy coefficient Hlxse of the draft of the supervision and law enforcement document through the data text correlation coefficient model and the data text legitimacy coefficient model, judge the legitimacy of the supervision and law enforcement document, form an internal supervision and law enforcement document, and send the internal supervision and law enforcement document to the database, wherein the correlation coefficient Glxse(w, i) of the text segment is calculated as follows: Where Glxse(w, i) is the correlation coefficient between text segment w and text segment i, z(w) is the subject dimension value of text segment w, z(i) is the subject dimension value of text segment i, y(w) is the semantic dimension value of text segment w, y(i) is the semantic dimension value of text segment i, k1, k2 are constants, and k1 = 0.2, k2 = 0.5; Among them, the calculation formula of the text legitimacy coefficient Hlxse is: Where Hlxse is the text legitimacy coefficient, Qa is the number of paragraphs in the supervision and law enforcement document, Ya is the completeness rate of the preset law enforcement event elements, Yb is the text sentence fluency index, r1, r2, r3, r4 are different preset proportional coefficients, and 0.1≤r1≤r2≤r3≤r4≤0.9, k is the adjustment factor; The artificial intelligence AI agent module is used to execute the artificial intelligence AI agent program, including calling external tools based on the collected health supervision and law enforcement event data information, generating external supervision and law enforcement documents, conducting joint analysis of internal supervision and law enforcement documents and external supervision and law enforcement documents, generating optimal supervision and law enforcement documents, and optimizing and storing supervision and law enforcement document models.

Images