Intelligent agent-based mine file content compliance automatic inspection method

Abstract

The invention provides a mine file content compliance automatic checking method based on an intelligent agent, and the method comprises the steps: determining a current to-be-checked key point and log information related to the to-be-checked key point, and the log information comprises historical task information and a historical execution result related to the to-be-checked key point; based on a first prompt template corresponding to a scheduling agent, inputting a to-be-checked key point and the log information into the scheduling agent to obtain first task information and a to-be-called target agent; based on the first task information, calling a target agent to obtain a first execution result returned by the target agent; updating the log information based on the first task information and the first execution result; and based on the updated log information, returning to execute the operation of determining the first task information and the to-be-called target agent until an inspection result of the to-be-inspected key point is obtained. According to the method, automatic inspection of the mine file is realized, the labor cost is saved, and the inspection efficiency and reliability are improved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing in the mining industry, and in particular to an agent-based automatic inspection method for compliance of mining file contents. Background Art

[0002] Mine production safety is undoubtedly the top priority of the entire mining industry. It is directly related to the life safety of employees, the stable operation of enterprises and the sustainable development of the environment. Therefore, for mine safety supervision agencies and mining enterprises themselves, regular compliance inspections of key documents such as various daily management records, production ledgers, equipment operation and maintenance logs of mining enterprises are an indispensable part of ensuring safe production.

[0003] Compliance inspection of documents usually requires inspectors to compare each piece of information recorded in the document (such as actual operations) with a large number of laws, regulations, national and industry standards and other compliance documents. This not only requires inspectors to have solid professional qualities, but also increases the difficulty and time-consuming nature of the inspection work due to the cumbersome, complex, and constantly updated laws and standards. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first objective of this application is to propose an agent-based automatic inspection method for compliance of mine document content.

[0006] The second objective of the present application is to propose an agent-based automatic inspection device for the compliance of mine document content.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth objective of the present application is to provide a computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above-mentioned purpose, the first embodiment of the present application proposes an agent-based automatic inspection method for compliance of mine file content, comprising: Determine the current key points to be checked and the log information associated with the key points to be checked, wherein the log information includes the historical task information and historical execution results associated with the key points to be checked; Based on the first prompt template corresponding to the scheduling agent, the key points to be checked and the log information are input into the scheduling agent to obtain the first task information and the target agent to be called; Based on the first task information, calling the target agent, and obtaining a first execution result returned by the target agent; Based on the first task information and the first execution result, updating the log information; Based on the updated log information, the operation of determining the first task information and the target agent to be called is returned until the inspection result of the key point to be inspected is obtained.

[0011] To achieve the above-mentioned purpose, the second embodiment of the present application proposes an agent-based automatic inspection device for compliance of mine file content, comprising: A determination module, used to determine the current key point to be checked and the log information associated with the key point to be checked, wherein the log information includes the historical task information and historical execution results associated with the key point to be checked; A processing module, for inputting the key points to be checked and the log information into the scheduling agent based on the first prompt template corresponding to the scheduling agent, and obtaining the first task information and the target agent to be called; A scheduling module, used to call a target agent based on the first task information and obtain a first execution result returned by the target agent; An updating module, used for updating the log information based on the first task information and the first execution result; The processing module is also used to return to the operation of determining the first task information and the target intelligent agent to be called based on the updated log information until the inspection result of the key point to be inspected is obtained.

[0012] To achieve the above-mentioned purpose, the third aspect of the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method as described in the embodiment of the first aspect.

[0013] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement the method described in the first aspect embodiment.

[0014] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0015] The subject-based automatic inspection method for compliance of mine file content provided by the present application, in the embodiment of the present application, firstly, the scheduling agent determines the current first task information of the key points to be inspected and the target agent to be called according to the key points to be inspected and the associated log information, and then obtains the current first execution result by calling the target agent, and then the scheduling agent determines the next task based on the updated log information, and so on, until the inspection result of the key points to be inspected is obtained. Thereby, the automatic inspection of the inspection points is realized, the labor cost is saved, and the efficiency and accuracy of the file inspection are improved. And the scheduling agent independently decides the next execution plan according to the actual situation, and calls the relevant agent to perform the corresponding task, so that the system has strong adaptability and robustness to the input file content.

[0016] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flowchart of an agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application; Figure 2 A schematic diagram of a scheduling process of a scheduling agent provided in an embodiment of the present application; Figure 3 A flowchart of another agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application; Figure 4 A schematic diagram of the retrieval process of the intelligent agent for searching for regulations provided in the embodiment of the present application; Figure 5 A schematic diagram of the inspection process of the content analysis inspection agent provided in an embodiment of the present application; Figure 6 A schematic diagram of a process for generating an inspection report provided in an embodiment of the present application; Figure 7 A schematic diagram of the overall process of an agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0019] The following describes an agent-based automatic inspection method for compliance of mine file content according to an embodiment of the present application with reference to the accompanying drawings.

[0020] Compliance inspection of mining industry documents requires inspectors to conduct a comprehensive and detailed review of the vast amount of documents to ensure that every detail complies with the regulations. This requires them to have not only solid professional knowledge, but also sufficient patience and care to cope with the heavy reading and analysis tasks.

[0021] On the other hand, compliance inspection is not a simple text comparison, but requires a one-by-one comparison of the actual operation of mining companies with a large number of laws, regulations, national and industry standards and other compliance documents. These laws and standards are often cumbersome and complex, and are constantly updated and changed, which undoubtedly increases the difficulty and time-consuming nature of the inspection work.

[0022] What is even more difficult is that even if the inspectors make great efforts, it is still difficult to completely avoid the risk of misjudgment and omission due to manual omissions and subjective understanding bias. This risk is very likely to lead to the omission of safety hazards.

[0023] In order to improve inspection efficiency and accuracy and liberate human resources to the greatest extent, the present invention provides an automatic inspection method for mine file content compliance based on intelligent agent, which can realize the automation and intelligence of the entire inspection link, greatly improving the efficiency and accuracy of mine file content compliance inspection.

[0024] Figure 1 A flowchart of an agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application.

[0025] like Figure 1 As shown, the agent-based automatic inspection method for compliance of mine file content includes: Step 101: determine the current key points to be checked and the log information associated with the key points to be checked.

[0026] The log information includes historical task information and historical execution results associated with the key points to be checked.

[0027] In some embodiments, the key points to be checked are used to describe the contents that need to be checked by the automatic inspection system, such as checking whether the operation and maintenance log of a certain operation or a certain device complies with laws and regulations, national and industry standards and regulations, etc.

[0028] In some embodiments, the inspection points can be input by the user. That is, the inspector can directly input the inspection points to be inspected into the automatic inspection system through the automatic inspection request. The inspector can input the inspection points to be inspected into the automatic inspection system through file upload or dialog box typing. After that, the automatic inspection system can determine the current inspection points from the received inspection request to automatically inspect the inspection points.

[0029] In some embodiments, the key points to be checked can also be determined by the automatic inspection system after parsing and splitting the document to be checked. The automatic inspection system can automatically match the format of the document to be checked, and use the corresponding technology to parse the document into a structured text format (such as a document in PDF format, which can be parsed by OCR technology); then, if there is a table in the file, the Large Language Model (LLM) can be used to convert the table content into text content described in natural language. Afterwards, the processed content to be checked can be split into several key points to be checked according to semantic units (usually split according to paragraphs) to form a list of key points to be checked.

[0030] In some embodiments, since there are many types of documents in the mining industry that need to be checked, the checkpoints corresponding to different types of documents may be different. In the present disclosure, different candidate checkpoint lists can be configured for different types of documents. When parsing the document to be checked, the candidate checkpoint list can be determined based on the type of the document to be checked, and then the document to be checked is parsed based on the candidate checkpoint list to determine the checkpoint list corresponding to the document to be checked. This improves the accuracy and reliability of the checkpoint list determined by the inspection system, and ensures the integrity and reliability of the file inspection.

[0031] In some embodiments, the automatic inspection system may include multiple agents, and different agents may complete different types of tasks. For example, the automatic inspection system may include a scheduling agent, which is used to plan the next task for the key points to be inspected; a specification and regulation retrieval agent, which can retrieve the currently required specifications and regulations according to the instructions of the scheduling agent; a content analysis inspection agent, which can inspect and analyze the key points to be inspected according to the specifications and regulations and determine the inspection results; and an inspection result recording agent, which can format and record the inspection results.

[0032] In some embodiments, since the inspection of the key points requires the use of relevant laws and regulations, different key points may rely on different laws and regulations, and the order or number of operations to be performed may be different. Therefore, the automatic inspection system can create log information of each key point when inspecting it, so as to provide an operation basis for the scheduling agent and other agents.

[0033] Step 102, based on the first prompt template corresponding to the scheduling agent, the key points to be checked and the log information are input into the scheduling agent to obtain the first task information and the target agent to be called.

[0034] In some embodiments, the automatic inspection system can use the key points to be inspected and the log information to fill in the unfilled positions in the first prompt template to obtain the corresponding prompt information, and then input the prompt information into the scheduling agent to obtain the first task information output by the scheduling agent and the target agent to be called.

[0035] In some embodiments, the first prompt template may include the following items to be filled in: LLM task items (such as generating the next execution plan based on the overall goal and the actions and results of the previous step in the log. The execution plan must be specific and targeted at a single specific task), decision basis (the decision depends on the content of the execution log, and the log can automatically accumulate the results and tasks of each round of execution), callable agents (including the specific functions and input parameter descriptions of the three agents: the regulation retrieval agent (or retrieval agent), the content analysis inspection agent, and the inspection result recording agent), and output content and format (the output format is consistent with the elements in the execution log to facilitate overall data flow management).

[0036] In some embodiments, the first task information may include the next task to be performed determined by the scheduling agent, the decision basis for the next task, the format and content of the result of the next task, etc.

[0037] In some embodiments, the target agent may be any one of a regulation retrieval agent, a content analysis inspection agent, and an inspection result recording agent.

[0038] Step 103: Based on the first task information, call the target agent to obtain the first execution result returned by the target agent.

[0039] In some embodiments, after determining the target intelligent agent, the automatic inspection system can first determine the second prompt template associated with the target intelligent agent, and then generate second prompt information based on the second prompt template corresponding to the target intelligent agent and the first task information; and then input the second prompt information into the target intelligent agent to obtain the first execution result output by the target intelligent agent.

[0040] In some embodiments, the target agent may be any one of a regulation retrieval agent, a content analysis inspection agent, and an inspection result recording agent.

[0041] Correspondingly, the first execution result may be a retrieval result, or a check result, or a record result, etc.

[0042] Step 104: Update the log information based on the first task information and the first execution result.

[0043] In some embodiments, the automatic inspection system can set a log storage area for the key points to be inspected to store information about the key points to be inspected during the inspection process, such as the tasks (or operations) performed, the execution results, etc., to provide an information source for each intelligent agent.

[0044] In some embodiments, the log information may include four elements: "feedback result of this step" (recording the execution result information fed back by the target agent in this step), "next step description (recording the next task information (action description) generated by the scheduler agent)", "next step execution node and execution parameters" (recording the target agent identifier and corresponding input parameter information required for the next action) and "retrieved and recalled related regulations set (cumulative)" (cumulatively recording the relevant regulations recalled for all retrieval behaviors during this inspection process).

[0045] In some embodiments, the contents in the log information may be stored therein in steps. For example, after a key point to be checked and an execution log (initial elements are empty) are input into the scheduler agent, the scheduler agent generates the next task information, the corresponding execution tool (target agent) and execution parameters. Afterwards, the content of the "next step description" part in the log information can be updated according to the content generated by the scheduling agent. Afterwards, according to the next execution tool and execution parameters in the execution log, the target agent (for example, one of the three agents, the specification and regulation retrieval agent, the content analysis and inspection agent, and the inspection result recording agent) is called to perform a specific task and obtain the execution result, so that the content of the "feedback result of this step" part in the log information can be updated.

[0046] In some embodiments, when the target agent is not an inspection result recording agent, it can be determined that the current first execution result is not the final inspection result corresponding to the key point to be inspected, that is, other tasks need to be performed for the current key point to be inspected to obtain the corresponding final inspection result. Therefore, the automatic inspection system can update the log information based on the first task information and the first execution result.

[0047] In some embodiments, if the target agent is an inspection result recording agent, that is, the first execution result is the final inspection result corresponding to the current key point to be inspected, then the automatic inspection system can store the first execution result and the key point to be inspected in association.

[0048] Step 105, based on the updated log information, return to the operation of determining the first task information and the target agent to be called, until the inspection result of the key point to be inspected is obtained.

[0049] In some embodiments, after obtaining the updated log information, the automatic inspection system can continue to input the updated log information and key points to be inspected into the scheduling agent, so that the scheduling agent can make scheduling decisions, determine the next task and the target agent to be called, etc.

[0050] Combine the following Figure 2 The scheduling process diagram of the scheduling agent shown in the figure takes a multi-round scheduling example as an example to further explain the scheduling process of the scheduling agent: First round: After determining the key points to be checked and the log information (which is empty at this time), the scheduling agent determines through scheduling decisions that the next step is to call the "regulations retrieval agent" to search for relevant evidence for the key points to be checked, and the agent returns the search results; Second round: Based on the updated log information, the scheduling agent decides to call the "content analysis and inspection agent" next, which automatically checks the key points to be checked based on the search results in the first round. The agent then gives feedback that "the basis for the judgment of xx aspects is missing"; Round 3: Based on the feedback from the second round, the scheduling agent calls the "regulations retrieval agent" again to search for relevant evidence for the missing xx aspects, and the agent returns the search results; Round 4: The scheduling agent decides on the next step to call the "content analysis and inspection agent" to automatically check the key points to be checked based on the search results in the first and third rounds, and feedback the inspection results. The content analysis and inspection agent feedbacks the inspection results, such as "xx aspects are illegal, the judgment basis is YY, and no violations are found in other aspects"; Round 5: The scheduling agent decides to call the “inspection result recording agent” in the next step to record the inspection results in a structured manner into a unified inspection result collection variable.

[0051] Round 6: The scheduling agent decides to end the task and the execution is completed.

[0052] In the embodiment of the present application, the scheduling agent first determines the current first task information of the key points to be checked and the target agent to be called according to the key points to be checked and the associated log information, and then obtains the current first execution result by calling the target agent. After that, the scheduling agent determines the next task based on the updated log information, and so on, until the inspection result of the key points to be checked is obtained. Thereby, the automatic inspection of the inspection points is realized, the labor cost is saved, and the efficiency and accuracy of the file inspection are improved. And the scheduling agent independently decides the next execution plan according to the actual situation, and calls the relevant agents to perform the corresponding tasks, so that the system has strong adaptability and robustness to the input file content.

[0053] Figure 3 A flowchart of another agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application.

[0054] like Figure 3 As shown, the agent-based automatic inspection method for compliance of mine file content includes: Step 301: parse the document to be checked and determine a list of checkpoints corresponding to the document to be checked.

[0055] Step 302: traverse the checkpoint list to determine the current checkpoint.

[0056] In some embodiments, after the user determines the content to be checked (such as the key points to be checked or the document to be checked), the user can first upload the content to be checked to the automatic inspection system. After obtaining the document to be checked, the automatic inspection system can parse the document, convert the table content into natural language, etc., to complete the splitting of the check points and obtain a list of check points. Then, the list of check points is traversed to complete the inspection of each check point.

[0057] Step 303: Based on the first prompt template corresponding to the scheduling agent, the key points to be checked are input into the scheduling agent to obtain the first task information and the target agent to be called.

[0058] The specific implementation of the above steps 301-303 can refer to the detailed description of steps 101-102 in the above embodiment, which will not be repeated here.

[0059] Step 304: Based on the first task information, call the target agent to obtain the first execution result returned by the target agent.

[0060] In some embodiments, if the target agent is a regulation retrieval agent, the process by which the target agent obtains the first execution result may be as shown in FIG. 4 . Figure 4 A schematic diagram of the retrieval process of the regulatory retrieval agent provided in the embodiment of the present application.

[0061] like Figure 4 As shown, after receiving the first task information, the regulation retrieval agent can first parse the first task information to determine the query statement (Query) and the target knowledge base (such as the mining laws and regulations database, or the mining standards and specifications database, etc.); then, according to the query statement, query the target knowledge base through dense retrieval and sparse retrieval respectively to obtain the first query result; then rerank the first query result (for example, rerank according to the semantic relevance between the first query result and the query statement) to obtain the first reranked result and the confidence of the first reranked result (for example, the confidence can be determined as the semantic relevance between the query statement); if the confidence of the first reranked result is greater than the threshold, then the first reranked result can be determined as the first execution result.

[0062] In some embodiments, if the confidence of the first rearrangement result is less than or equal to a threshold, the retrieval agent can first determine the keywords corresponding to the query statement (for example, extract keywords from the query statement using the LLM model); then query the target knowledge base based on the keywords to obtain a second query result; and then rearrange the second query result (for example, rearrange according to the semantic relevance between the second query result and the query statement) to obtain the first execution result.

[0063] In some embodiments, in order to improve the retrieval efficiency of the retrieval agent, such as Figure 4 As shown in the figure, the automatic inspection system can first collect various mining laws and regulations and national and industry standards and other normative documents; then parse the documents using professional technologies (such as optical character recognition (OCR)), and split the text blocks according to the file structure; then use the embedded model to vectorize the semantics of each text block, and use sparse retrieval models (such as BM25, BM42, etc.) to build a sparse index; finally, store the text block and structured information (such as hierarchical structure) and attachments (such as tables, formulas) into a vector database to form a knowledge base.

[0064] In some embodiments, if the target agent is a content analysis and inspection agent (Agent), the process of obtaining the first execution result can be as follows: Figure 5 shown. Figure 5 A schematic diagram of the inspection process of the content analysis inspection agent provided in an embodiment of the present application.

[0065] like Figure 5As shown in the figure, the main function of the content analysis and inspection agent is to analyze and inspect the key points to be inspected based on the retrieved reference regulations. Its core is an LLM analyzer, which is essentially an LLM after prompt engineering matching. It can use the powerful reasoning ability of the large language model to compare and analyze the key points to be inspected and the reference regulations, and generate a conclusion.

[0066] The LLM analyzer analyzes and checks the key points to be checked and the relevant regulations for retrieval and recall to determine whether some key reference information is still missing. If so, it generates a description of the missing information; otherwise, it generates the inspection results of the key points to be checked.

[0067] Step 305: Update the log information based on the first task information and the first execution result.

[0068] Step 306, based on the updated log information, return to the operation of determining the first task information and the target agent to be called, until the inspection result of the key point to be inspected is obtained.

[0069] In some embodiments, when the target agent determined after the scheduling agent makes a decision is the inspection result recording agent, the inspection result recording agent can structure each inspection result (such as dividing it into inspection conclusions, inspection details, inspection basis, reasons for judgment, etc.) and output it to a unified inspection result collection variable for storage.

[0070] The specific implementation of the above steps 305-306 can refer to the detailed description of steps 104-105 in the above embodiment, which will not be repeated here.

[0071] Step 307: When all the check points in the check point list have been checked, an inspection report corresponding to the document to be checked is generated according to the inspection results corresponding to all the check points in the check point list.

[0072] Since a document to be inspected may correspond to multiple inspection points, in order to improve the readability of the inspection results, the automatic inspection system can summarize and organize all inspection results to generate a highly readable and rich inspection report. Its execution process can be as follows: Figure 6 shown. Figure 6 A schematic diagram of the process of generating an inspection report provided in an embodiment of the present application. First, a preset report template (prepared in advance) is selected according to the inspection document, and then the numerical statistics of key information (such as the number of compliance, the number of non-compliance, the compliance ratio, etc.) are performed using statistical analysis techniques according to the inspection results. Furthermore, statistical charts can also be generated (for example, using Text2Code technology), and finally an inspection report is generated according to the template.

[0073] From the above analysis, it can be seen that the overall process of the automatic inspection method for the compliance of mine file content based on the intelligent agent provided by the present disclosure can be as follows: Figure 7 shown. Figure 7 A schematic diagram of the overall process of an agent-based automatic inspection method for compliance of mine file content provided in an embodiment of the present application.

[0074] The agent-based automatic inspection method for compliance of mine file content in the embodiment of the present application realizes automatic inspection of compliance of mine file content by utilizing the coordination and cooperation of multiple agents. It reduces labor costs and improves inspection efficiency and accuracy. In addition, it uses a comprehensive dense + sparse + full-text retrieval and Reranker rearrangement filtering mechanism to ensure high accuracy and completeness of retrieval recall, avoid erroneous recall and omissions, and is of great significance for compliance inspection. In addition, by leveraging the powerful reasoning ability of the large language model, it completes the automated analysis, comparison and judgment of the key points to be inspected, realizes automation and intelligence of file content inspection, and avoids misjudgments caused by human own quality. Finally, a rich and highly readable inspection report is output to facilitate users to clearly grasp key information.

[0075] In order to implement the above-mentioned embodiment, the present application also proposes an automatic inspection device for the compliance of mine file contents based on an intelligent agent.

[0076] The agent-based automatic inspection device for compliance of mine file contents includes: A determination module, used to determine the current key point to be checked and the log information associated with the key point to be checked, wherein the log information includes the historical task information and historical execution results associated with the key point to be checked; A processing module, for inputting the key points to be checked and the log information into the scheduling agent based on the first prompt template corresponding to the scheduling agent, and obtaining the first task information and the target agent to be called; A scheduling module, used to call a target agent based on the first task information and obtain a first execution result returned by the target agent; An updating module, used for updating the log information based on the first task information and the first execution result; The processing module is also used to return to the operation of determining the first task information and the target intelligent agent to be called based on the updated log information until the inspection result of the key point to be inspected is obtained.

[0077] Furthermore, in a possible implementation of the embodiment of the present application, the update module is used to: In the case where the target agent is not the inspection result recording agent, the log information is updated based on the first task information and the first execution result.

[0078] Furthermore, in a possible implementation of the embodiment of the present application, a storage module is also included, which is used to associate and store the first execution result and the key points to be checked when the target agent is an inspection result recording agent.

[0079] Furthermore, in a possible implementation of the embodiment of the present application, the scheduling module is further configured to: Generate second prompt information based on the second prompt template corresponding to the target intelligent agent and the first task information; The second prompt information is input into the target intelligent agent to obtain a first execution result output by the target intelligent agent.

[0080] Furthermore, in a possible implementation of the embodiment of the present application, the target agent is a retrieval agent, and the processing module is further used to: Parsing the first task information to determine the query statement and the target knowledge base; According to the query statement, query the target knowledge base through dense search and sparse search respectively to obtain a first query result; Rearranging the first query result to obtain a first rearranged result and a confidence level of the first rearranged result; When the confidence level of the first rearrangement result is greater than a threshold, the first rearrangement result is determined to be the first execution result.

[0081] Furthermore, in a possible implementation of the embodiment of the present application, the processing module is further configured to: When the confidence level of the first rearrangement result is less than or equal to the threshold, determining the keyword corresponding to the query statement; Based on the keywords, query the target knowledge base to obtain a second query result; The second query result is rearranged to obtain the first execution result.

[0082] Furthermore, in a possible implementation manner of the embodiment of the present application, the above-mentioned determination module is also used to: obtain the current key points to be inspected from the received inspection request.

[0083] Furthermore, in a possible implementation of the embodiment of the present application, the above-mentioned determination module is also used to: Parse the documents to be checked and determine the list of checkpoints corresponding to the documents to be checked; Go through the checkpoint list to determine the current checkpoint.

[0084] Furthermore, in a possible implementation of the embodiment of the present application, the above-mentioned determination module is also used to: Determine a list of candidate checkpoints based on the type of document to be checked; Based on the candidate checkpoint list, the document to be checked is parsed to determine the checkpoint list corresponding to the document to be checked.

[0085] Furthermore, in a possible implementation of the embodiment of the present application, the processing module is further configured to: When all the check points in the check point list have been checked, an inspection report corresponding to the document to be checked is generated according to the inspection results corresponding to all the check points in the check point list.

[0086] It should be noted that the above explanation of the embodiment of the agent-based automatic inspection method for compliance of mine file contents is also applicable to the agent-based automatic inspection method device for compliance of mine file contents in this embodiment, and will not be repeated here.

[0087] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments. In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0088] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0089] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they contradict each other.

[0090] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0091] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0092] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0093] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0094] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0095] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0096] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An agent-based automatic inspection method for compliance of mine file content, characterized in that: The following steps are involved: Determine the current key points to be checked and the log information associated with the key points to be checked, wherein the log information includes the historical task information and historical execution results associated with the key points to be checked; Based on the first prompt template corresponding to the scheduling agent, the key points to be checked and the log information are input into the scheduling agent to obtain the first task information and the target agent to be called; Based on the first task information, calling the target agent to obtain a first execution result returned by the target agent; Based on the first task information and the first execution result, updating the log information; Based on the updated log information, the operation of determining the first task information and the target agent to be called is returned to be executed until the inspection result of the key point to be inspected is obtained.

2. The method according to claim 1, characterized in that The updating of the log information based on the first task information and the first execution result includes: In the case where the target agent is not an inspection result recording agent, the log information is updated based on the first task information and the first execution result.

3. The method according to claim 1, characterized in that After obtaining the first execution result returned by the target agent, the method further includes: In the case where the target agent is an inspection result recording agent, the first execution result and the key points to be inspected are stored in association.

4. The method according to claim 1, characterized in that The step of calling the target agent based on the first task information to obtain a first execution result returned by the target agent includes: Generate second prompt information based on a second prompt template corresponding to the target agent and the first task information; The second prompt information is input into the target agent to obtain a first execution result output by the target agent.

5. The method according to claim 1, characterized in that The target agent is a retrieval agent, and the process of the target agent obtaining the first execution result includes: Parsing the first task information to determine a query statement and a target knowledge base; According to the query statement, query the target knowledge base through dense search and sparse search respectively to obtain a first query result; Rearranging the first query result to obtain a first rearranged result and a confidence level of the first rearranged result; When the confidence level of the first rearrangement result is greater than a threshold, the first rearrangement result is determined as a first execution result.

6. The method according to claim 5, characterized in that After obtaining the first rearrangement result and the confidence level of the first rearrangement result, the method further includes: When the confidence level of the first rearrangement result is less than or equal to a threshold, determining a keyword corresponding to the query statement; Based on the keyword, query the target knowledge base to obtain a second query result; The second query result is rearranged to obtain a first execution result.

7. The method according to any one of claims 1 to 6, characterized in that: The determination of the current key points to be checked includes: Get the current points to be inspected from the received inspection request.

8. The method according to any one of claims 1 to 6, characterized in that: The determination of the current key points to be checked includes: Parsing the document to be checked to determine a list of key points to be checked for the document to be checked; The checkpoint list is traversed to determine the current checkpoint to be checked.

9. The method according to claim 8, characterized in that The step of parsing the document to be checked and determining a list of checkpoints corresponding to the document to be checked includes: Determining a list of candidate inspection points according to the type of the document to be inspected; Based on the candidate checkpoint list, the document to be checked is parsed to determine a checkpoint list corresponding to the document to be checked.

10. The method according to claim 8, characterized in that The method further comprises: In the case that all the check points in the check point list have been checked, an inspection report corresponding to the document to be checked is generated according to the inspection results corresponding to all the check points in the check point list.

Images