Recommendation method for intelligent audit and compliance analysis of assets

Abstract

The invention discloses a recommendation method for intelligent audit and compliance analysis of assets, and relates to the technical field of asset management, audit analysis and compliance evaluation, and the method comprises the steps: obtaining asset data related to asset management from a plurality of pieces of source data of an enterprise, and carrying out the cleaning and preprocessing of the asset data, and obtaining the asset data with consistent quality; performing auditing analysis on the asset data by using a machine learning algorithm, identifying asset use efficiency, maintenance cost abnormity and potential auditing risks, and obtaining an auditing analysis result; and based on laws and regulations and enterprise management specifications, performing compliance analysis by using a natural language processing technology to obtain an asset compliance analysis result. According to the invention, an enterprise can realize automation and intelligentization of asset management, so that the accuracy and compliance of auditing are improved, the operation risk is reduced, and the use and maintenance strategies of assets are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of asset management, audit analysis and compliance assessment, and in particular to a recommendation method for intelligent asset auditing and compliance analysis. Background Art

[0002] Auditing and compliance analysis are crucial components of asset management to ensure that asset use and management adhere to laws, regulations, and corporate governance practices. As businesses expand and diversify, asset management becomes increasingly complex. Traditional auditing methods face challenges such as inefficiency, inaccuracy, and difficulty adapting to rapidly evolving compliance requirements.

[0003] Current asset audit processes largely rely on manual processes, which are not only time-consuming and labor-intensive but also susceptible to subjective judgment, leading to biased audit results. Furthermore, compliance analysis requires specialized knowledge and a deep understanding of regulations to ensure companies remain compliant in an ever-changing legal landscape. However, with the continuous refinement of regulations, companies face tremendous pressure in compliance management.

[0004] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0005] In response to the problems in the related technologies, the present invention proposes a recommended method for asset intelligent auditing and compliance analysis to overcome the above-mentioned technical problems existing in the existing related technologies.

[0006] To this end, the specific technical solutions adopted in the present invention are as follows: A recommended approach for asset intelligence auditing and compliance analysis, including: Acquire asset data related to asset management from multiple source data of the enterprise, and clean and pre-process the asset data to obtain asset data of consistent quality; Use machine learning algorithms to audit and analyze asset data, identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks, and obtain audit analysis results; Based on laws, regulations and corporate management standards, and using natural language processing technology to conduct compliance analysis, we can obtain asset compliance analysis results; Provide customized improvement suggestions and action plans to asset managers based on audit analysis results and compliance analysis results; Leverage learning mechanisms to adaptively optimize new data and user feedback, and dynamically adjust audit and compliance analysis standards.

[0007] Furthermore, asset data related to asset management is obtained from multiple source data of the enterprise, and the asset data is cleaned and preprocessed to obtain asset data of consistent quality, including: Based on multiple data sources of the enterprise, define data source sets and query the same asset data through the database to ensure automatic acquisition and preliminary preparation of asset data; Use data cleaning rules to remove duplicates, correct formats, and fill missing values ​​in the synchronized asset data to obtain cleaned asset data; Based on consistency check rules, asset data in different data sources are checked for consistency in logic, value, time, and format to ensure that asset data remains accurate after multi-source integration; Based on the asset data after cleaning and consistency testing, weighted fusion is used to process the numerical attributes of the asset data, and the categorical attributes of the asset data are processed through a majority voting mechanism to obtain asset data with consistent quality.

[0008] Furthermore, the formula for weighted fusion is: ; Where, S Indicates the result value of the fused numerical attribute; oh i Indicates the i The weight of each data source; c i Indicates the i The confidence level of each data source; Indicates the i Numeric attribute values ​​from a data source that have been cleaned, formatted, and preprocessed; n Indicates the total number of data sources involved in the fusion.

[0009] Furthermore, we use machine learning algorithms to conduct audit analysis on asset data to identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks. The audit analysis results include: Based on asset management data, a feature matrix is ​​constructed and standardized to obtain feature data of uniform scale; Use feature engineering methods to extract key nonlinear features from asset data and reduce data dimensions; Build an anomaly detection model and use the extracted key nonlinear features to train the anomaly detection model, calculate the reconstruction error of each key nonlinear feature, and determine the anomaly threshold; Based on the asset utilization efficiency index, the time series analysis method is used to predict the asset utilization trend and obtain the asset efficiency classification results; Determine whether the asset maintenance cost is abnormal based on maintenance cost indicators and cost-benefit ratio; Combining maintenance cost and usage efficiency, a risk scoring function is defined, the risk score of the asset is calculated, and the identification results of potential audit risks are obtained.

[0010] Furthermore, based on the asset utilization efficiency index, the time series analysis method is used to predict the asset utilization trend, and the asset efficiency classification results are as follows: Based on the asset usage data, obtain the monthly usage of the year, perform time series analysis, and obtain the annual usage benchmark value of the asset; Using the annual usage benchmark value, calculate the monthly efficiency status ratio to obtain the efficiency status of the asset, and then classify the efficiency by comparing the efficiency status with the annual usage benchmark value; Based on the obtained efficiency status ratio, a three-level classification standard is established, and assets are divided into high-efficiency, normal and low-efficiency categories to obtain the efficiency classification results of assets.

[0011] Furthermore, the calculation formula for the annual usage benchmark value of the asset is: ; Where, B Indicates the annual usage base value of the asset; U t Indicates the monthly usage.

[0012] Furthermore, the formula for calculating the monthly efficiency status ratio is: ; Where, S t Indicates the efficiency status ratio per month; U t Indicates the monthly usage; B Indicates the annual usage baseline value of the asset.

[0013] Furthermore, based on laws, regulations, and corporate management standards, and using natural language processing technology to conduct compliance analysis, the asset compliance analysis results include: Build a regulatory knowledge base, collect text collections of laws, regulations, and corporate management specifications related to asset management, and pre-process the text collections to obtain structured regulatory text data; Use natural language processing technology to extract keywords that represent compliance rules from legal and regulatory texts; Treat compliance rules as a series of conditional statements, define the compliance standards for asset use and management through each conditional statement, and assign a weight to each compliance rule; The asset management data is treated as a set of feature vectors and converted into feature vectors based on data characterization to ensure that each asset management feature vector can fully describe its usage and management status; Based on the weight of the compliance rules and the matching degree of the asset feature vector, a compliance scoring model is constructed to calculate the compliance score for each asset feature vector to obtain the compliance score of each asset feature vector; Set a compliance score threshold. If the compliance score of an asset feature vector is less than the threshold, it is judged as non-compliant. Otherwise, it is judged as compliant. For each asset feature vector, its compliance score and decision result are output to generate a compliance report.

[0014] Furthermore, a regulatory knowledge base is constructed to collect text collections of laws, regulations, and corporate management specifications related to asset management. The text collections are preprocessed to obtain structured regulatory text data including: Build a legal knowledge base to collect text collections of laws, regulations, and corporate management specifications related to asset management. Remove redundant characters, unify the encoding, and standardize the upper and lower case of the text collection to obtain cleaned legal and regulatory texts. Using natural language processing technology, we segment the cleaned legal and regulatory texts, optimize the segmentation of professional terms, and remove common and domain stop words to obtain high-quality legal and regulatory word segmentation results. Based on the obtained word segmentation results of laws and regulations, part-of-speech tagging tools are used to perform grammatical attribute annotation, extract key phrases and semantic information, and obtain structured legal text data.

[0015] Furthermore, the formula for calculating the compliance score for each asset feature vector is: ; Where, m i Indicates the i The weight of the rule; ( r i , f ) represents the asset feature vector f and rules r i degree of matching.

[0016] The beneficial effects of the present invention are: 1. The present invention improves data quality by integrating multi-source asset data and performing cleaning processing; combines machine learning with natural language processing to achieve intelligent identification of asset utilization efficiency, maintenance cost anomalies and compliance issues; provides customized improvement suggestions based on audit and compliance analysis results; dynamically optimizes analysis standards through learning mechanisms, enhances the adaptability and accuracy of the system, and comprehensively improves the intelligence and compliance level of asset management, enabling enterprises to achieve automation and intelligence in asset management, thereby improving audit accuracy and compliance, reducing operational risks, and optimizing asset use and maintenance strategies.

[0017] 2. This invention improves data accuracy and integrity through multi-source asset data cleaning and consistency testing; adopts weighted fusion and majority voting mechanisms to effectively resolve data conflicts and heterogeneity problems, ensuring the uniformity and reliability of asset data quality; provides high-quality data support for subsequent intelligent auditing and compliance analysis, and enhances the accuracy of analysis results and decision-making support capabilities.

[0018] 3. The present invention accurately extracts asset usage trends through time series analysis based on asset utilization efficiency indicators; combines efficiency status ratios and annual benchmark values ​​to scientifically classify assets into high-efficiency, normal, and low-efficiency categories, thereby improving the accuracy of asset operation status identification; and provides strong support for asset management optimization and resource allocation decision-making, thereby improving management efficiency.

[0019] 4. The present invention extracts compliance rules by constructing a regulatory knowledge base and applying natural language processing technology; combines asset feature vectors with rule matching to quantify asset compliance levels and accurately identify non-compliant assets; and automatically generates compliance analysis reports to improve the compliance detection efficiency and decision-making support capabilities of asset management, thereby assisting enterprises in risk control. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. 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 any creative work.

[0021] Figure 1 The present invention is a flowchart of a method for intelligent asset auditing and compliance analysis. DETAILED DESCRIPTION

[0022] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and the advantages of the present invention.

[0023] According to an embodiment of the present invention, a recommended method for intelligent asset auditing and compliance analysis is provided.

[0024] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, the recommended method for intelligent asset auditing and compliance analysis according to an embodiment of the present invention includes: S1. Obtain asset data related to asset management from multiple source data of the enterprise, and clean and pre-process the asset data to obtain asset data of consistent quality; Specifically, it automatically collects asset management-related data from multiple data sources across the enterprise, including but not limited to asset acquisition records, depreciation information, maintenance logs, usage status, and historical transactions. It also provides data cleaning and pre-processing capabilities to ensure the quality and consistency of input data.

[0025] S2. Use machine learning algorithms to conduct audit analysis on asset data, identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks, and obtain audit analysis results; Specifically, machine learning algorithms are used to conduct in-depth analysis of collected asset data to identify asset utilization efficiency, maintenance costs, and potential audit risks. This enables automatic detection of data anomalies and inconsistencies, providing a basis for subsequent compliance analysis.

[0026] S3. Based on laws, regulations, and corporate management standards, and using natural language processing technology, conduct compliance analysis to obtain asset compliance analysis results; Specifically, we assess whether the use and management of assets meet prescribed compliance standards, combining the latest laws, regulations, and corporate management standards. Through natural language processing technology, we can understand and interpret complex regulatory texts and apply them to actual asset management activities.

[0027] S4. Provide customized improvement suggestions and action plans to asset managers based on audit analysis results and compliance analysis results; Specifically, based on the results of intelligent audits and compliance analysis, asset managers are provided with customized improvement recommendations and action plans. These recommendations may include asset reallocation, maintenance plan adjustments, and compliance risk mitigation measures. User feedback is also taken into account to continuously optimize the relevance and accuracy of recommendations.

[0028] S5. Leverage learning mechanisms to adaptively optimize new data and user feedback, and dynamically adjust audit and compliance analysis standards.

[0029] Specifically, through self-learning and optimization mechanisms, the standards for audit and compliance analysis can be continuously adjusted based on new data and user feedback, ensuring the continued effectiveness and adaptability of recommended methods.

[0030] In this optional embodiment, asset data related to asset management is obtained from multiple source data of the enterprise, and the asset data is cleaned and preprocessed to obtain asset data of consistent quality, including: Based on multiple data sources of the enterprise, define data source sets and query the same asset data through the database to ensure automatic acquisition and preliminary preparation of asset data; Use data cleaning rules to remove duplicates, correct formats, and fill missing values ​​in the synchronized asset data to obtain cleaned asset data; Based on consistency check rules, asset data in different data sources are checked for consistency in logic, value, time, and format to ensure that asset data remains accurate after multi-source integration; Based on the asset data after cleaning and consistency testing, weighted fusion is used to process the numerical attributes of the asset data, and the categorical attributes of the asset data are processed through a majority voting mechanism to obtain asset data with consistent quality.

[0031] Specifically, the multi-source data automatic synchronization and preprocessing algorithm includes: 1) Data source identification and synchronization: Define a data source collection D ={ D 1, D 2, ..., D n},in D i Representative i data sources.

[0032] Periodically synchronize data from each data source using API calls or database queries.

[0033] 2) Data cleaning: For each data source D i , define data cleaning rules R i , including removing duplicate records, correcting format errors, filling missing values, etc.

[0034] Applying Rules R i to data D i , get the cleaned data , the formula is: .

[0035] 3) Data consistency check: Design consistency rules, which are a set of predefined logical constraints that ensure consistency of identical or related data across different data sources in terms of logic, numerical values, time, and other dimensions. The goal is to automatically detect and resolve conflicts between multi-source data, thereby ensuring the accuracy and reliability of the fused data.

[0036] Field value consistency rules: the key fields of the same entity should be completely consistent in different data sources.

[0037] Logical consistency rules: the logical relationship between data should conform to business rules or mathematical formulas.

[0038] Time consistency rules: time-related data should conform to time sequence or timeliness.

[0039] Format consistency rules: the format of the same field should be unified across all data sources.

[0040] Inconsistent data is resolved based on predefined priority rules or through manual review.

[0041] If data conflicts can be resolved using predefined rules (such as data source authority or timestamp priority), prioritize them. This approach is suitable for data conflicts where a clear judgment can be made, such as when certain data sources are more trustworthy or certain fields have stricter formatting rules.

[0042] If data conflicts cannot be resolved through priority rules or require more complex judgments (such as those involving subjective judgments or requiring further verification), they are resolved through manual review.

[0043] 4) Data Fusion: For numerical attributes (such as asset valuation and maintenance cost), the weighted fusion formula is: .

[0044] Where, S Indicates the result value of the fused numerical attribute, which represents the final value after weighted calculation of multiple data sources; oh i Indicates the i The weight of a data source indicates the credibility or importance of the data source. The higher the weight, the greater the contribution to the final result. c i Indicates the i The confidence level of a data source, reflecting the quality or reliability of the data source; Indicates the i Numeric attribute values ​​from a data source that have been cleaned, formatted, and preprocessed; n Indicates the total number of data sources involved in the fusion.

[0045] For categorical attributes in asset management (such as asset status, equipment type, etc.), data conflicts are resolved through multi-source data fusion. oh i ) and confidence ( c i), a dynamically adjusted majority voting mechanism is designed to ensure that the fusion results reflect both the consistency of the majority of data sources and the authority of high-weight and high-confidence data sources.

[0046] From each data source D i Extract the cleaned and formatted categorical attribute values v i ,verify v i Whether it conforms to the predefined set of legal values V , illegal values ​​are marked as “unknown” and excluded from voting.

[0047] Calculate the voting weight of each data source. W i By its basic weight oh i and confidence c i Jointly decided: .

[0048] Perform weighted majority voting on each candidate value v ∈ V , calculate its total voting weight and select the highest candidate value as the final result: .

[0049] In this optional embodiment, a machine learning algorithm is used to perform audit analysis on asset data to identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks. The audit analysis results include: Based on asset management data, a feature matrix is ​​constructed and standardized to obtain feature data of uniform scale; Use feature engineering methods to extract key nonlinear features from asset data and reduce data dimensions; Build an anomaly detection model and use the extracted key nonlinear features to train the anomaly detection model, calculate the reconstruction error of each key nonlinear feature, and determine the anomaly threshold; Based on the asset utilization efficiency index, the time series analysis method is used to predict the asset utilization trend and obtain the asset efficiency classification results; Determine whether the asset maintenance cost is abnormal based on maintenance cost indicators and cost-benefit ratio; Combining maintenance cost and usage efficiency, a risk scoring function is defined, the risk score of the asset is calculated, and the identification results of potential audit risks are obtained.

[0050] In this optional embodiment, based on the asset utilization efficiency index, a time series analysis method is used to predict the asset utilization trend, and the asset efficiency classification results obtained include: Based on the asset usage data, obtain the monthly usage of the year, perform time series analysis, and obtain the annual usage benchmark value of the asset; Using the annual usage benchmark value, calculate the monthly efficiency status ratio to obtain the efficiency status of the asset, and then classify the efficiency by comparing the efficiency status with the annual usage benchmark value; Based on the obtained efficiency status ratio, a three-level classification standard is established, and assets are divided into high-efficiency, normal and low-efficiency categories to obtain the efficiency classification results of assets.

[0051] Specifically, the intelligent asset management audit algorithm based on anomaly detection includes: 1) Data representation: Representing asset management data as a feature matrix X ∈ ,in m It is the number of data points (asset records), such as frequency of use, maintenance costs, etc.

[0052] Normalize the input data to eliminate dimensional differences.

[0053] 2) Feature Engineering: By applying feature engineering methods, an autoencoder is used to extract key nonlinear features from asset management data and reduce the data dimension, providing high-information feature representation for subsequent anomaly detection and risk assessment.

[0054] 3) Anomaly Detection Model: Train an anomaly detection model such as Isolation Forest, One-Class SVM, or Autoencoder Use the high-information feature matrix output by feature engineering X ∈ (such as dimensionality reduction features extracted by autoencoders).

[0055] Standardize the feature matrix (such as Z-score standardization) to ensure stable model convergence.

[0056] Training data: The model is trained using only normal data (unlabeled) so that the model learns the distribution pattern of normal data.

[0057] Calculate the reconstruction error: For each sample X i , calculate the MSE of its input and output.

[0058] Determine the abnormal threshold based on the reconstruction error distribution of normal samples (such as the 95% quantile) i .

[0059] Model Evaluation: Precision: The proportion of abnormal samples correctly identified.

[0060] Recall: The proportion of correctly identified abnormal samples in actual samples.

[0061] Use a labeled test set (containing known anomaly samples) to verify model performance.

[0062] 4) Usage efficiency analysis: Define usage efficiency metrics, such as the ratio of how often an asset is used to how often it is expected to be used.

[0063] Use time series analysis to comprehensively predict asset usage trends and identify inefficient assets: Data preparation: Collect assets recently t Period (default t = 12 months) (monthly usage for the year): .

[0064] Where, U Indicates the total amount of usage of the asset; U t Indicates the monthly usage.

[0065] To calculate the annual benchmark value (the annual usage benchmark value of the asset), expand the benchmark value into a 12-month horizontal straight line series. The formula for calculating the annual benchmark value (the annual usage benchmark value of the asset) is: .

[0066] Where, B Indicates the annual usage base value of the asset; U t Indicates the monthly usage.

[0067] Efficiency registration classification: Calculate efficiency status ratio S t , establish a three-level classification standard, calculate the efficiency status ratio each month (calculate the efficiency status ratio S t ) is: .

[0068] Where, S tIndicates the efficiency status ratio per month; U t Indicates the monthly usage; B Indicates the annual usage baseline value of the asset.

[0069] High efficiency (high efficiency category): efficiency status ≥80%.

[0070] Normal (normal class): 50% ≤ efficiency status < 80%.

[0071] Inefficiency (inefficiency category): efficiency status <50%.

[0072] 5) Maintenance cost analysis: Defining maintenance cost metrics M , such as the ratio of an asset's maintenance costs to its value.

[0073] Use cost-benefit analysis to determine whether the cost of maintaining an asset is reasonable.

[0074] Input data: Maintenance cost data: single maintenance cost, annual maintenance total cost (from maintenance log).

[0075] Asset value data: original value, current valuation, and residual value (derived from asset acquisition records and depreciation information).

[0076] Key indicator calculations: Maintenance cost ratio ( M ): .

[0077] Cost-effectiveness ratio ( CBR ): .

[0078] Reasonable range determination: If the current assets M If the range is beyond ±2 times the standard deviation of the cluster center, it will be marked as abnormal.

[0079] Reasonable settings CBR Threshold (such as CBR ≤1 means benefit ≥ cost, which is considered reasonable).

[0080] like CBR >1, it is judged as “maintenance cost is not economical”.

[0081] 6) Risk Assessment: Define the risk scoring function R ( x ), which calculates risk scores based on asset utilization efficiency, maintenance costs, and other relevant factors.

[0082] Design a threshold i , and assign risk scores higher than i Assets are flagged as potential audit risks.

[0083] In this optional embodiment, based on laws, regulations, and enterprise management specifications, and utilizing natural language processing technology to perform compliance analysis, the asset compliance analysis results include: Build a regulatory knowledge base, collect text collections of laws, regulations, and corporate management specifications related to asset management, and pre-process the text collections to obtain structured regulatory text data; Use natural language processing technology to extract keywords that represent compliance rules from legal and regulatory texts; Treat compliance rules as a series of conditional statements, define the compliance standards for asset use and management through each conditional statement, and assign a weight to each compliance rule; The asset management data is treated as a set of feature vectors and converted into feature vectors based on data characterization to ensure that each asset management feature vector can fully describe its usage and management status; Based on the weight of the compliance rules and the matching degree of the asset feature vector, a compliance scoring model is constructed to calculate the compliance score for each asset feature vector to obtain the compliance score of each asset feature vector; Set a compliance score threshold. If the compliance score of an asset feature vector is less than the threshold, it is judged as non-compliant. Otherwise, it is judged as compliant. For each asset feature vector, its compliance score and decision result are output to generate a compliance report.

[0084] In this optional embodiment, a regulatory knowledge base is constructed to collect a text collection of laws, regulations, and enterprise management specifications related to asset management. The text collection is preprocessed to obtain structured regulatory text data including: Build a legal knowledge base to collect text collections of laws, regulations, and corporate management specifications related to asset management. Remove redundant characters, unify the encoding, and standardize the upper and lower case of the text collection to obtain cleaned legal and regulatory texts. Using natural language processing technology, we segment the cleaned legal and regulatory texts, optimize the segmentation of professional terms, and remove common and domain stop words to obtain high-quality legal and regulatory word segmentation results. Based on the obtained word segmentation results of laws and regulations, part-of-speech tagging tools are used to perform grammatical attribute annotation, extract key phrases and semantic information, and obtain structured legal text data.

[0085] Specific compliance analysis algorithms based on rules and machine learning: 1) Construction of regulatory knowledge base: Create a regulatory knowledge base that contains a collection of texts of laws, regulations, and enterprise management norms related to asset management.

[0086] 2) Text preprocessing: Preprocess the texts in the regulatory knowledge base, including natural language processing such as word segmentation, stop word removal, and词性标注 (POS tagging). Remove redundant characters, delete HTML tags, special symbols (such as ";"), and illegal characters.

[0087] Unify encoding and case, convert the text to UTF-8 encoding uniformly, and standardize it to lowercase (in English scenarios) or preserve the case (in Chinese scenarios).

[0088] Use open-source tools (such as jieba), combined with a custom dictionary to optimize the segmentation of professional terms.

[0089] General stop word list (such as NLTK's stopwords) + domain-specific stop words (such as redundant words in regulations like "Article", "clause", "According to", etc.).

[0090] Use the POS tagging function of LTP, HanLP, or jieba.

[0091] 3) Feature extraction: Keyword extraction, extract keywords from the regulatory text that can represent compliance rules.

[0092] Phrase and clause number extraction, extract phrases and clause numbers from the regulatory text.

[0093] 4) Representation of compliance rules: Represent the compliance rules as a series of conditional statements, where each conditional statement defines the compliance criteria for asset use and management.

[0094] Represent the rules in the form of "condition → action", for example: Condition: The depreciation period of the asset exceeds 10 years.

[0095] Action: Asset scrapping must be carried out.

[0096] Assign weights to each rule, and the weights can be determined according to the priority of the regulations or historical audit data.

[0097] 5) Representation of asset management data: Represent the asset management data as a collection of feature vectors, where each vector x i Describes the use and management situation of an asset.

[0098] Data characterization, convert the asset management data into the form of feature vectors, including: Defines a collection of characteristics for asset management data, such as: Asset ID, acquisition date, depreciation information, maintenance records.

[0099] How often the asset is used, how much it costs to maintain it, and its current condition.

[0100] Map asset management data to feature vectors to ensure that the feature vector of each asset can fully describe its usage and management.

[0101] 6) Compliance Scoring Model: Construct a compliance scoring model S:X->[0,1], which is a feature vector for each asset management x i Calculate a compliance score.

[0102] The formula for defining the compliance score (calculated for each asset feature vector) is: .

[0103] Where, m i Indicates the i The weight of the rule; ( r i , f ) represents the asset feature vector f and rules r i degree of matching.

[0104] 7) Compliance Decisions: A compliance threshold τ is set. If the compliance score of an asset is less than τ, the asset management activity is considered non-compliant.

[0105] For each asset, output its compliance score and decision result (compliant / non-compliant).

[0106] Generate compliance reports detailing the reasons for violations and improvement recommendations for non-compliant assets.

[0107] In summary, with the help of the above technical solutions of the present invention, the present invention improves data accuracy and completeness through multi-source asset data cleaning and consistency detection; adopts weighted fusion and majority voting mechanisms to effectively solve data conflicts and heterogeneity problems, ensuring uniform and reliable asset data quality; provides high-quality data support for subsequent intelligent auditing and compliance analysis, and enhances the accuracy of analysis results and decision-making support capabilities. The present invention accurately extracts asset usage trends through time series analysis based on asset utilization efficiency indicators; scientifically divides assets into efficient, normal and inefficient categories based on efficiency status ratios and annual benchmark values, and improves the accuracy of asset operation status identification; provides strong support for asset management optimization and resource allocation decisions, and improves management efficiency. The present invention extracts compliance rules by constructing a legal knowledge base and applying natural language processing technology; combines asset feature vectors with rule matching to quantify asset compliance levels and accurately identify non-compliant assets; automatically generates compliance analysis reports, improves the compliance detection efficiency and decision-making support capabilities of asset management, and assists enterprise risk control.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any 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 recommended method for intelligent asset auditing and compliance analysis, characterized in that: include: Acquire asset data related to asset management from multiple source data of the enterprise, and clean and pre-process the asset data to obtain asset data of consistent quality; Use machine learning algorithms to audit and analyze asset data, identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks, and obtain audit analysis results; Based on laws, regulations and corporate management standards, and using natural language processing technology to conduct compliance analysis, we can obtain asset compliance analysis results; Provide customized improvement suggestions and action plans to asset managers based on audit analysis results and compliance analysis results; Leverage learning mechanisms to adaptively optimize new data and user feedback, and dynamically adjust audit and compliance analysis standards.

2. The method for recommending intelligent asset auditing and compliance analysis according to claim 1, characterized in that: The acquisition of asset data related to asset management from multiple source data of the enterprise, and the cleaning and pre-processing of the asset data to obtain asset data of consistent quality include: Based on multiple data sources of the enterprise, define data source sets and query the same asset data through the database to ensure automatic acquisition and preliminary preparation of asset data; Use data cleaning rules to remove duplicates, correct formats, and fill missing values ​​in the synchronized asset data to obtain cleaned asset data; Based on consistency check rules, asset data in different data sources are checked for consistency in logic, value, time, and format to ensure that asset data remains accurate after multi-source integration; Based on the asset data after cleaning and consistency testing, weighted fusion is used to process the numerical attributes of the asset data, and the categorical attributes of the asset data are processed through a majority voting mechanism to obtain asset data with consistent quality.

3. The method for recommending intelligent asset auditing and compliance analysis according to claim 2, characterized in that: The formula for weighted fusion is: ； Where, S Indicates the result value of the fused numerical attribute; ω i Indicates the i The weight of each data source; c i Indicates the i The confidence level of each data source; Indicates the i Numeric attribute values ​​from a data source that have been cleaned, formatted, and preprocessed; n Indicates the total number of data sources involved in the fusion.

4. The method for recommending intelligent asset auditing and compliance analysis according to claim 1, characterized in that: The audit analysis of asset data using machine learning algorithms can identify asset utilization efficiency, maintenance cost anomalies, and potential audit risks. The audit analysis results include: Based on asset management data, a feature matrix is ​​constructed and standardized to obtain feature data of uniform scale; Use feature engineering methods to extract key nonlinear features from asset data and reduce data dimensions; Build an anomaly detection model and use the extracted key nonlinear features to train the anomaly detection model, calculate the reconstruction error of each key nonlinear feature, and determine the anomaly threshold; Based on the asset utilization efficiency index, the time series analysis method is used to predict the asset utilization trend and obtain the asset efficiency classification results; Determine whether the asset maintenance cost is abnormal based on maintenance cost indicators and cost-benefit ratio; Combining maintenance cost and usage efficiency, a risk scoring function is defined, the risk score of the asset is calculated, and the identification results of potential audit risks are obtained.

5. The method for recommending intelligent asset auditing and compliance analysis according to claim 4, characterized in that: The asset efficiency index based on the asset is used to predict the asset usage trend using the time series analysis method, and the asset efficiency classification results obtained include: Based on the asset usage data, obtain the monthly usage of the year, perform time series analysis, and obtain the annual usage benchmark value of the asset; Using the annual usage benchmark value, calculate the monthly efficiency status ratio to obtain the efficiency status of the asset, and then classify the efficiency by comparing the efficiency status with the annual usage benchmark value; Based on the obtained efficiency status ratio, a three-level classification standard is established, and assets are divided into high-efficiency, normal and low-efficiency categories to obtain the efficiency classification results of assets.

6. The method for recommending intelligent asset auditing and compliance analysis according to claim 5, characterized in that: The calculation formula for the annual usage benchmark value of the asset is: ； Where, B Indicates the annual usage base value of the asset; U t Indicates the monthly usage.

7. The method for recommending intelligent asset auditing and compliance analysis according to claim 5, characterized in that: The formula for calculating the monthly efficiency status ratio is: ； Where, S t Indicates the efficiency status ratio per month; U t Indicates the monthly usage; B Indicates the annual usage baseline value of the asset.

8. The method for recommending intelligent asset auditing and compliance analysis according to claim 1, characterized in that: The asset compliance analysis results obtained by using natural language processing technology based on laws, regulations and corporate management standards include: Build a regulatory knowledge base, collect text collections of laws, regulations, and corporate management specifications related to asset management, and pre-process the text collections to obtain structured regulatory text data; Use natural language processing technology to extract keywords that represent compliance rules from legal and regulatory texts; Treat compliance rules as a series of conditional statements, define the compliance standards for asset use and management through each conditional statement, and assign a weight to each compliance rule; The asset management data is treated as a set of feature vectors and converted into feature vectors based on data characterization to ensure that each asset management feature vector can fully describe its usage and management status; Based on the weight of the compliance rules and the matching degree of the asset feature vector, a compliance scoring model is constructed to calculate the compliance score for each asset feature vector to obtain the compliance score of each asset feature vector; Set a compliance score threshold. If the compliance score of an asset feature vector is less than the threshold, it is judged as non-compliant. Otherwise, it is judged as compliant. For each asset feature vector, its compliance score and decision result are output and a compliance report is generated.

9. The method for recommending intelligent asset auditing and compliance analysis according to claim 8, characterized in that: The regulatory knowledge base is constructed by collecting a text collection of laws, regulations, and enterprise management specifications related to asset management, and preprocessing the text collection to obtain structured regulatory text data including: Build a legal knowledge base to collect text collections of laws, regulations, and corporate management specifications related to asset management. Remove redundant characters, unify the encoding, and standardize the upper and lower case of the text collection to obtain cleaned legal and regulatory texts. Using natural language processing technology, we segment the cleaned legal and regulatory texts, optimize the segmentation of professional terms, and remove common and domain stop words to obtain high-quality legal and regulatory word segmentation results. Based on the obtained word segmentation results of laws and regulations, part-of-speech tagging tools are used to perform grammatical attribute annotation, extract key phrases and semantic information, and obtain structured legal text data.

10. The method for recommending intelligent asset auditing and compliance analysis according to claim 8, characterized in that: The formula for calculating the compliance score for each asset feature vector is: ； Where, μ i Indicates the i The weight of the rule; ( r i , f ) represents the asset feature vector f and rules r i degree of matching.

Images