Data pre-auditing method based on risk supervision assistant

Abstract

The invention relates to the technical field of electric power, in particular to a data pre-auditing method based on a risk supervision assistant, which comprises the following steps: S1, a user submits data to be audited through a system interface, and the system performs format verification and preliminary screening on the data to ensure the integrity and compliance of the data; and the risk supervision assistant performs preliminary risk assessment on the data. Compared with the prior art, the method has the advantages that through a method of combining intellectualization and manpower, the powerful intellectualization processing capacity of the artificial intelligence technology is ingeniously fused, key information can be efficiently extracted from mass data, preliminary screening and classification are conducted, meanwhile, professional knowledge and experience judgment of an artificial auditor are fully utilized, and the intelligent processing capacity of the artificial intelligence technology is greatly improved. Especially, when complex, fuzzy or edge cases are processed, the intervention of manual auditing can ensure the accuracy and compliance of the auditing result, and the combination not only improves the auditing efficiency, but also significantly enhances the auditing efficiency.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a data pre-audit method based on a risk supervision assistant. Background Art

[0002] The main purpose of data review in the power industry is to ensure that the operating data of power companies are accurate, complete and compliant, thereby ensuring the safe, stable and efficient operation of the power industry. Through the review, errors and non-compliance in the data can be discovered and corrected in a timely manner, thereby improving the operating level and competitiveness of power companies.

[0003] In the process of data review in the power industry, manual review requires checking the data item by item, and the processing speed is relatively slow. Especially when faced with a large amount of data, the progress of manual review may lag significantly, resulting in an extension of the review cycle, affecting the overall efficiency of the review work. At the same time, during the manual review process, due to limited human attention and energy, facing a large amount of data and documents for a long time, auditors may become fatigued and negligent, resulting in errors in the review results.

[0004] Therefore, this application provides a document pre-review method based on a risk supervision assistant to meet the pre-review of submitted documents. Summary of the invention

[0005] In view of this, the purpose of the present invention is to propose a data pre-review method based on a risk supervision assistant to solve the problem that manual review of data is inefficient and prone to errors.

[0006] Based on the above purpose, the present invention provides a data pre-review method based on a risk supervision assistant, which specifically includes the following steps: S1: Users submit information for review through the system interface. The system performs format verification and preliminary screening on the information to ensure its integrity and compliance. The risk supervision assistant conducts a preliminary risk assessment on the data and identifies potential risk points based on preset risk rules and algorithms; S2: Use artificial intelligence technology to extract and compare key information in the data, and use natural language processing technology to perform semantic analysis on text data to extract key information; Compare the extracted key information with the preset standard library or database to verify the accuracy and compliance of the information; For information with risks or anomalies, the system automatically marks and generates a pre-review report, prompting manual further review; S3: The manual auditor reviews the marked risk points based on the pre-audit report generated by the system; Auditors use tools provided by the system, such as magnifying glasses and marking pens, to magnify and mark details in the data in order to more accurately identify risks; S4: Based on the results of manual assisted review, the system processes the data. If the data is compliant and accurate, it will be directly approved. If the data has risks or is non-compliant, it needs to be returned to the user for modification or supplementation. The system generates a detailed audit report, including pre-audit results, manual audit results, and modification suggestions, so that users can understand the audit process and improvement directions; The user modifies and improves the information based on the audit report and resubmits it for review. The system then conducts another pre-review and manual assisted review of the modified information until the information meets compliance and accuracy requirements.

[0007] Preferably, in S1, when the user submits information, the system provides classification options and labeling functions to help the user quickly locate the information type and add relevant labels for subsequent processing and retrieval. Before formal submission, the system provides a preview function to allow the user to check the completeness and accuracy of the information to avoid repeated submission due to misoperation or omissions.

[0008] Preferably, the system has a built-in format verification module to perform format checks on the submitted materials, including file size, resolution, and encoding format, to ensure that the materials meet the system processing requirements. For materials that do not meet the format requirements, the system gives a clear error prompt and guides the user to make corrections.

[0009] Preferably, the key information extraction comprises the following steps: S211: Collect the text data to be processed, convert the collected data into a computer-processable format, remove irrelevant characters in the text, and ensure that the text content is clean and neat; S212: Divide the text into meaningful words or phrases, determine the part of speech of each word in the sentence, which helps in subsequent information extraction, identifies entities in the text, analyzes the dependency relationship between words in the sentence, and helps in understanding the structure and semantics of the sentence.

[0010] S213: Extract specific key information from the text according to preset rules or patterns, use the trained model to automatically identify and extract key information through machine learning algorithms, and combine rule-based and machine learning-based methods to improve the accuracy and generalization ability of information extraction.

[0011] Preferably, the key information comparison comprises the following steps: S221: Collect and organize standard information related to the text to be processed to ensure that the data format in the standard library is consistent with the format of the extracted key information; S222: directly compare the extracted key information with the information in the standard library to determine whether they are completely consistent. For information with spelling errors or format differences, use a fuzzy matching algorithm to compare them, and use semantic analysis technology to compare whether the semantics of the two sentences are similar or the same; S223: For information with inconsistent comparison results or questionable information, the system automatically marks it as risky or abnormal information.

[0012] Preferably, in S3, for the parts of the document with smaller fonts and blurred details, the auditor uses a magnifying glass tool provided by the system to zoom in and view them, and the magnifying glass tool supports different magnifications.

[0013] Preferably, in S3, the auditor uses a marking pen tool to mark key details in the document, and the marking pen tool provides a variety of colors and shapes to distinguish different types of markings.

[0014] Preferably, in S4, the system generates a detailed audit report, the content of which covers the preliminary audit results, manual audit results and modification suggestions. The preliminary audit results part includes the risk points and potential problems initially judged by the system; the manual audit results part records in detail the manual auditor's review process, problems found and handling suggestions.

[0015] Preferably, in S4, the audit report uses clear and concise language to avoid excessive use of professional terms so that users can easily understand the audit process and results. Key information in the report is visually prompted by bold and color to facilitate users to quickly locate and understand.

[0016] Preferably, in S4, during the cyclic review process, the system continuously optimizes the pre-review algorithm and the manual review process. At the same time, the system collects and analyzes user feedback and continuously improves the review standards and the generation mechanism of modification suggestions.

[0017] Beneficial effects of the present invention: 1. This data pre-review method based on the risk supervision assistant, through the combination of intelligent and manual methods, has demonstrated outstanding advantages in the current complex and changeable data review environment. It cleverly integrates the powerful intelligent processing capabilities of artificial intelligence technology, such as natural language processing, image recognition, and big data analysis, and can efficiently extract key information from massive data for preliminary screening and classification. At the same time, it makes full use of the professional knowledge and experience judgment of manual auditors. Especially when dealing with complex, ambiguous or marginal cases, the intervention of manual review can ensure the accuracy and compliance of the audit results. This combination not only improves the audit efficiency, but also significantly enhances the comprehensiveness and reliability of the audit, forming an audit system with complementary advantages and mutual support, providing double protection for data quality and security.

[0018] 2. This document pre-review method based on the risk supervision assistant, through the application of the risk supervision assistant, traditionally, document review often relies on manual experience and limited data analysis, and it is difficult to fully capture potential risks. The risk supervision assistant integrates advanced algorithm models and data analysis tools to realize intelligent pre-review of large amounts of data. It can automatically identify abnormal patterns, warn of potential risks, and generate preliminary risk assessment reports. What's more advanced is that it can also seamlessly connect with the manual review process, and through manual assisted review, the results of the intelligent pre-review are reviewed and optimized, thereby ensuring the accuracy and depth of the document review, and greatly improving the efficiency and precision of the document review. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or 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 in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 The present invention provides a schematic diagram of the steps of a risk supervision assistant-based data pre-review method. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0022] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0023] like Figure 1 As shown, a data pre-review method based on a risk supervision assistant specifically includes the following steps: S1: Users submit information for review through the system interface. The system performs format verification and preliminary screening on the information to ensure its integrity and compliance. The risk supervision assistant conducts a preliminary risk assessment on the data and identifies potential risk points based on preset risk rules and algorithms; In S1, when users submit materials, the system provides classification options and labeling functions to help users quickly locate the material type and add relevant labels for subsequent processing and retrieval. Before formal submission, the system provides a preview function to allow users to check the completeness and accuracy of the materials to avoid repeated submission due to misoperation or omissions; The system has a built-in format verification module to check the format of the submitted materials, including file size, resolution, and encoding format, to ensure that the materials meet the system processing requirements. For materials that do not meet the format requirements, the system will give clear error prompts and guide users to make corrections; S2: Use artificial intelligence technology to extract and compare key information in the data, and use natural language processing technology to perform semantic analysis on text data to extract key information; Compare the extracted key information with the preset standard library or database to verify the accuracy and compliance of the information; For information with risks or anomalies, the system automatically marks and generates a pre-review report, prompting manual further review; Key information extraction includes the following steps: S211: Collect the text data to be processed, convert the collected data into a computer-processable format, remove irrelevant characters in the text, and ensure that the text content is clean and neat; S212: Divide the text into meaningful words or phrases, determine the part of speech of each word in the sentence, which helps in subsequent information extraction, identifies entities in the text, analyzes the dependency relationship between words in the sentence, and helps in understanding the structure and semantics of the sentence.

[0024] S213: Extract specific key information from the text according to preset rules or patterns, use the trained model to automatically identify and extract key information through machine learning algorithms, and combine rule-based and machine learning-based methods to improve the accuracy and generalization ability of information extraction; Key information comparison includes the following steps: S221: Collect and organize standard information related to the text to be processed to ensure that the data format in the standard library is consistent with the format of the extracted key information; S222: directly compare the extracted key information with the information in the standard library to determine whether they are completely consistent. For information with spelling errors or format differences, use a fuzzy matching algorithm to compare them, and use semantic analysis technology to compare whether the semantics of the two sentences are similar or the same; S223: For information with inconsistent or questionable comparison results, the system automatically marks it as risky or abnormal information; S3: The manual auditor reviews the marked risk points based on the pre-audit report generated by the system; Auditors use tools provided by the system, such as magnifying glasses and marking pens, to magnify and mark details in the data in order to more accurately identify risks; In S3, for parts of the document with small fonts and blurred details, auditors use the magnifying glass tool provided by the system to zoom in and view them. The magnifying glass tool supports different magnifications; In S3, auditors use the annotation pen tool to annotate key details in the data. The annotation pen tool provides multiple colors and shapes to distinguish different types of annotations; S4: Based on the results of manual assisted review, the system processes the data. If the data is compliant and accurate, it will be directly approved. If the data has risks or is non-compliant, it needs to be returned to the user for modification or supplementation. The system generates a detailed audit report, including pre-audit results, manual audit results, and modification suggestions, so that users can understand the audit process and improvement directions; The user modifies and improves the information based on the audit report and resubmits it for review. The system conducts another pre-review and manual assisted review on the modified information until the information meets the compliance and accuracy requirements; In S4, the system generates a detailed audit report, which includes the pre-audit results, manual audit results, and modification suggestions. The pre-audit results include the risk points and potential problems initially determined by the system; the manual audit results record in detail the manual auditor's review process, problems found, and handling suggestions; In S4, the audit report uses clear and concise language, avoiding excessive use of professional terms, so that users can easily understand the audit process and results. The key information in the report is bolded and colored to help users quickly locate and understand it; In S4, during the cyclic review process, the system continuously optimizes the pre-review algorithm and manual review process. At the same time, the system collects and analyzes user feedback and continuously improves the review standards and modification suggestion generation mechanism.

[0025] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0026] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data pre-review method based on risk supervision assistant, It is characterized in that The specific steps include: S1: Users submit information for review through the system interface. The system performs format verification and preliminary screening on the information to ensure its integrity and compliance. The risk supervision assistant conducts a preliminary risk assessment on the data and identifies potential risk points based on preset risk rules and algorithms; S2: Use artificial intelligence technology to extract and compare key information in the data, and use natural language processing technology to perform semantic analysis on text data to extract key information; Compare the extracted key information with the preset standard library or database to verify the accuracy and compliance of the information; For information with risks or anomalies, the system automatically marks and generates a pre-review report, prompting manual further review; S3: The manual auditor reviews the marked risk points based on the pre-audit report generated by the system; Auditors use tools provided by the system, such as magnifying glasses and marking pens, to magnify and mark details in the data in order to more accurately identify risks; S4: Based on the results of manual assisted review, the system processes the data. If the data is compliant and accurate, it will be directly approved. If the data has risks or is non-compliant, it needs to be returned to the user for modification or supplementation. The system generates a detailed audit report, including pre-audit results, manual audit results, and modification suggestions, so that users can understand the audit process and improvement directions; The user modifies and improves the information based on the audit report and resubmits it for review. The system then conducts another pre-review and manual assisted review of the modified information until the information meets compliance and accuracy requirements.

2. According to the method for pre-reviewing information based on the risk supervision assistant of claim 1, It is characterized in that In S1, when the user submits information, the system provides classification options and labeling functions to help the user quickly locate the information type and add relevant labels for subsequent processing and retrieval. Before formal submission, the system provides a preview function to allow the user to check the completeness and accuracy of the information to avoid repeated submission due to misoperation or omissions.

3. According to the method for pre-reviewing information based on the risk supervision assistant of claim 1, It is characterized in that The system has a built-in format verification module to check the format of the submitted materials, including file size, resolution, and encoding format, to ensure that the materials meet the system processing requirements. For materials that do not meet the format requirements, the system will give clear error prompts and guide users to make corrections.

4. According to the method for pre-reviewing data based on the risk supervision assistant of claim 1, It is characterized in that The key information extraction comprises the following steps: S211: Collect the text data to be processed, convert the collected data into a computer-processable format, remove irrelevant characters in the text, and ensure that the text content is clean and neat; S212: Divide the text into meaningful words or phrases, determine the part of speech of each word in the sentence, which is helpful for subsequent information extraction, identifying entities in the text, analyzing the dependency relationship between words in the sentence, and helping to understand the structure and semantics of the sentence; S213: Extract specific key information from the text according to preset rules or patterns, use the trained model to automatically identify and extract key information through machine learning algorithms, and combine rule-based and machine learning-based methods to improve the accuracy and generalization ability of information extraction.

5. According to the method for pre-reviewing information based on the risk supervision assistant of claim 4, It is characterized in that The key information comparison includes the following steps: S221: Collect and organize standard information related to the text to be processed to ensure that the data format in the standard library is consistent with the format of the extracted key information; S222: directly compare the extracted key information with the information in the standard library to determine whether they are completely consistent. For information with spelling errors or format differences, use a fuzzy matching algorithm to compare them, and use semantic analysis technology to compare whether the semantics of the two sentences are similar or the same; S223: For information with inconsistent comparison results or questionable information, the system automatically marks it as risky or abnormal information.

6. According to claim 1, a method for pre-reviewing information based on a risk supervision assistant, It is characterized in that In S3, for the parts of the data with smaller fonts and blurred details, the auditor uses the magnifying glass tool provided by the system to zoom in and view them. The magnifying glass tool supports different magnifications.

7. According to claim 1, a data pre-review method based on a risk supervision assistant, It is characterized in that In S3, the auditor uses a marking pen tool to mark key details in the data. The marking pen tool provides a variety of colors and shapes to distinguish different types of markings.

8. According to claim 1, a method for pre-reviewing information based on a risk supervision assistant, It is characterized in that In S4, the system generates a detailed audit report, which covers the preliminary audit results, manual audit results and modification suggestions. The preliminary audit results include the risk points and potential problems initially determined by the system; the manual audit results record in detail the manual auditor's review process, problems found and handling suggestions.

9. According to claim 1, a method for pre-reviewing information based on a risk supervision assistant, It is characterized in that In S4, the audit report uses clear and concise language and avoids excessive use of professional terms so that users can easily understand the audit process and results. Key information in the report is highlighted in bold and colored for visual cues so that users can quickly locate and understand it.

10. According to claim 1, a method for pre-reviewing information based on a risk supervision assistant, It is characterized in that In S4, during the cyclic review process, the system continuously optimizes the pre-review algorithm and manual review process. At the same time, the system collects and analyzes user feedback and continuously improves the review standards and modification suggestion generation mechanism.

Images