CN111126461B - Intelligent auditing method based on machine learning model interpretation - Google Patents
Intelligent auditing method based on machine learning model interpretation Download PDFInfo
- Publication number
- CN111126461B CN111126461B CN201911272684.XA CN201911272684A CN111126461B CN 111126461 B CN111126461 B CN 111126461B CN 201911272684 A CN201911272684 A CN 201911272684A CN 111126461 B CN111126461 B CN 111126461B
- Authority
- CN
- China
- Prior art keywords
- audit
- machine learning
- learning model
- auditing
- rule
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the field of intelligent audit, in particular to an intelligent audit method based on machine learning model interpretation, aiming at the problems that the existing audit method mostly adopts a manual mode to audit, the rule and rule to be followed in the audit process are updated quickly, the range is large, the accuracy and the reliability of the audit can be influenced only by the brain storage of auditors, and the risk of the audit is high, the following scheme is provided: s1: inputting a file or data to be audited into an input end of an intelligent audit system based on machine learning model interpretation; s2: auditing the input file or data and simultaneously displaying the auditing process by using a machine learning model; s3: the machine learning result is visualized and the business rule is output through the model interpretation function of the machine learning; the invention is beneficial to improving the accuracy and reliability of audit and can perfect the life cycle management of audit rules.
Description
Technical Field
The invention relates to the technical field of intelligent audit, in particular to an intelligent audit method based on machine learning model interpretation.
Background
Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. The computer is specially researched how to simulate or realize the learning behavior of human beings so as to acquire new knowledge or skills, and the existing knowledge structure is reorganized to continuously improve the performance of the computer, so that the machine learning model can be well integrated in intelligent audit.
Most of the existing auditing methods adopt a manual mode to audit, rules and rules to be followed in the auditing process are updated quickly, the range is wide, the auditing accuracy and reliability can be affected only by the storage of the brain of an auditor, and the auditing risk is high, so that an intelligent auditing method based on machine learning model interpretation is provided, and the problems are solved.
Disclosure of Invention
The intelligent auditing method based on machine learning model interpretation solves the problems that the existing auditing method mostly adopts a manual mode to audit, rules and rules to be followed in the auditing process are updated quickly, the range is wide, the accuracy and reliability of auditing can be affected only by the brain storage of auditors, and the risk of auditing is high.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the intelligent auditing method based on machine learning model interpretation comprises the following steps:
s1: inputting a file or data to be audited into an input end of an intelligent audit system based on machine learning model interpretation;
s2: auditing the input file or data and simultaneously displaying the auditing process by using a machine learning model;
s3: the machine learning result is visualized and the business rule is output through the model interpretation function of the machine learning;
s4: comparing the output business rule with the original audit rule base, adding the newly added business rule into the audit rule base, and updating the audit rule base;
s5: deleting the rules in the updated audit rule base, and performing single discarding on two rules with similarity exceeding eighty percent.
Preferably, in the S1, the smart audit system explained based on the machine learning model has a four-layer two-wing structure, and the data layer, the information layer, the knowledge layer and the smart layer are respectively arranged from bottom to top, and the left wing is an information technology supporting part, which is beneficial to regulations and rules.
Preferably, in the step S2, the input file or data is audited in a decision tree manner in the auditing process, rules used in the auditing process are added in the internal memory of the machine learning model one by one and the details of the rules are stored together, and the final result of the audit is displayed on the machine learning model.
Preferably, in the step S3, an Ante-hoc interpretability is used in the machine learning, so that a decision process or a decision basis of the model can be understood without additional information, and the operations are completed on SQL, and then the machine learning model result is displayed on a display screen.
Preferably, in the step S3, the business rule output is to output the rules stored in the audit process in the step S2 one by one, and a counter is further arranged on a memory inside the machine learning model in the step S2, so that all the used rules can be summarized.
Preferably, the main function of the data layer is to store and share numeric data and non-numeric data required by audit work, and the main function of the intelligent layer is to generate and share accurate judgment and suggestion of audit service national governance.
Preferably, the information layer has a main function of generating and sharing the relation contained in the data, the information layer can represent the source and the destination of the data, and the information technology support part has a main function of providing information technology support guarantee for mutual rotation among the data layer, the information layer, the knowledge layer and the intelligent layer.
Preferably, in the step S4, the municipal audit rule base is searched before updating the audit rule base, the newly added and modified rule is stored at the update layer, the business rule used in the audit process is compared with the audit rule base, the audit rule is repeated and omitted, and the newly added or modified audit rule at the update layer is added into the audit rule base.
Preferably, in the step S5, the rules in the updated audit rule base are compared by using a global comparison algorithm, and the rules with high similarity are deleted from the rules with no update in the late stage.
Preferably, in the step S5, an updating component is further provided in the audit rule base, the updating component records the service time of each rule, and each updating component traverses the audit rule base once, so as to delete the unused rule within half a year, and the purpose is to perfect the life cycle of the audit rule.
Compared with the prior art, the invention has the beneficial effects that:
1. the technical scheme of the invention mainly solves the problems that the auditing method adopts a manual mode to audit, rules and rules to be followed in the auditing process are updated quickly, the range is large, the auditing accuracy and reliability can be influenced only by the brain storage of auditors, and the auditing risk is high;
2. the intelligent auditing method based on machine learning model interpretation can continuously perfect the machine learning model, can reversely supplement the newly added or modified business rules to the auditing rule base, is beneficial to improving the accuracy and reliability of auditing, and can perfect the life cycle management of the auditing rules;
3. the invention can realize intelligent audit, can provide support and help for auditors, greatly reduces the working intensity of the auditors, is beneficial to improving the accuracy and reliability of audit, and can perfect the life cycle management of audit rules.
Drawings
Fig. 1 is a flowchart of a smart audit method based on machine learning model interpretation according to the present invention.
Detailed Description
The following description of the technical solutions in the embodiments of the present invention will be clear and complete, and it is obvious that the described embodiments are only some embodiments of the present invention, but not all embodiments.
Examples
Referring to fig. 1, a smart audit method based on machine learning model interpretation includes the steps of:
s1: inputting a file or data to be audited into an input end of an intelligent audit system based on machine learning model interpretation;
s2: auditing the input file or data and simultaneously displaying the auditing process by using a machine learning model;
s3: the machine learning result is visualized and the business rule is output through the model interpretation function of the machine learning;
s4: comparing the output business rule with the original audit rule base, adding the newly added business rule into the audit rule base, and updating the audit rule base;
s5: deleting the rules in the updated audit rule base, and performing single discarding on two rules with similarity exceeding eighty percent.
In the embodiment, in S1, the intelligent audit system based on machine learning model interpretation has four layers of two wing structures, namely a data layer, an information layer, a knowledge layer and an intelligent layer from bottom to top, the left wing is an information technology supporting part, which is beneficial to laws and regulations and rules, in S2, the input files or data are audited in the audit process in a decision tree mode, rules used in the audit process are added in the internal memory of the machine learning model one by one together with details of the storage rules, the final result of the audit is displayed on the machine learning model, in S3, the Ante-hoc interpretability is used in the machine learning, the decision process or decision basis of the model can be understood without additional information, the operations are completed on SQL, and then the result of the machine learning model is displayed on a display screen, in S3, the business rule output is to output the rules stored in the audit process in S2 one by one, a counter is also arranged on a memory in the machine learning model in S2, all the used rules can be summarized, the main function of the data layer is to store and share numerical data and non-numerical data required by audit work, the main function of the intelligent layer is to generate and share accurate judgment and suggestion of audit service national governance, the main function of the information layer is to generate and share relations contained in the data, the information layer can represent the source and the destination of the data, the main function of the information technology support part is to provide information technology support guarantee for mutual inversion among the data layer, the information layer, the knowledge layer and the intelligent layer, in S4, the municipal audit rule library is searched before the audit rule library is updated, and the new and modified rules are stored at the update layer, comparing business rules used in the auditing process with an auditing rule base, checking for duplicate and mending leakage, adding the newly added or modified auditing rules at the updating layer into the auditing rule base, comparing the rules in the updated auditing rule base by adopting a global comparison algorithm, deleting the rules with high similarity, and deleting the rules which are not updated in the recent period, wherein an updating component is further arranged in the auditing rule base, the updating component records the service time of each rule, each traversing the auditing rule base for one time, and deleting the rules which are not used in half a year, so as to perfect the life cycle of the auditing rules.
The invention can realize intelligent audit, can provide support and help for auditors, greatly reduces the working intensity of the auditors, is beneficial to improving the accuracy and reliability of audit, and can perfect the life cycle management of audit rules.
The technical scheme of the invention solves the problems that the existing auditing method mostly adopts a manual mode to audit, the rule and rule to be followed in the auditing process are updated quickly, the range is large, the accuracy and reliability of the auditing can be influenced only by the brain storage of the auditor, and the risk of the auditing is high.
The present invention is not limited to the above-mentioned embodiments, and any person skilled in the art, based on the technical solution of the present invention and the inventive concept thereof, can be replaced or changed within the scope of the present invention.
Claims (6)
1. The intelligent auditing method based on the machine learning model interpretation is characterized by comprising the following steps of:
s1: inputting a file or data to be audited into an input end of an intelligent audit system based on machine learning model interpretation; in the S1, the intelligent audit system based on machine learning model interpretation is provided with four layers of two-wing structures, namely a data layer, an information layer, a knowledge layer and an intelligent layer from bottom to top, and the left wing is an information technology supporting part, and is beneficial to a regulation and rule part; the main function of the data layer is to store and share the numerical data and the non-numerical data required by the audit work, and the main function of the intelligent layer is to generate and share the accurate judgment and suggestion of the audit;
s2: auditing the input file or data and simultaneously displaying the auditing process by using a machine learning model; in the step S2, the input file or data is audited in a decision tree mode in the auditing process, rules used in the auditing process are added in the internal memory of the machine learning model one by one to store details of the rules, and the final result of the audit is displayed on the machine learning model;
s3: the machine learning result is visualized and the business rule is output through the model interpretation function of the machine learning;
s4: comparing the output business rule with the original audit rule base, adding the newly added business rule into the audit rule base, and updating the audit rule base; in the step S4, the municipal audit rule base is searched before the audit rule base is updated, the newly added and modified rule is stored at the update layer, the business rule used in the audit process is compared with the audit rule base, the audit rule base is searched for duplicate and leakage, and the newly added or modified audit rule at the update layer is added into the audit rule base;
s5: deleting the rules in the updated audit rule base, and performing single discarding on two rules with similarity exceeding eighty percent.
2. The intelligent auditing method based on machine learning model interpretation according to claim 1, wherein in S3, the Ante-hoc interpretability is used in the machine learning, the decision process or decision basis of the model can be understood without additional information, the operations are all completed on SQL, and then the machine learning model result is displayed on a display screen.
3. The intelligent auditing method based on machine learning model interpretation according to claim 1, characterized in that in S3, the business rule output is to output the rules stored in the auditing process in S2 one by one, and a counter is further arranged on the memory inside the machine learning model in S2, so that all the used rules can be summarized.
4. The intelligent audit method based on machine learning model interpretation according to claim 1, wherein the information layer has a main function of generating and sharing the relationship contained in the data, the information layer can represent the source and destination of the data, and the information technology support part has a main function of providing information technology support guarantee for mutual rotation among the layers of the data layer, the information layer, the knowledge layer and the intelligent layer.
5. The intelligent auditing method based on machine learning model interpretation according to claim 1, characterized in that in S5, the rules in the updated auditing rule base are compared by using a global comparison algorithm, and the rules with high similarity are deleted.
6. The intelligent auditing method based on machine learning model interpretation according to claim 1, wherein in S5, an updating component is further set in the auditing rule base, the updating component records the service time of each rule, each updating component traverses the auditing rule base once, and the unused rule in half a year is deleted, so as to perfect the life cycle of the auditing rule.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911272684.XA CN111126461B (en) | 2019-12-12 | 2019-12-12 | Intelligent auditing method based on machine learning model interpretation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911272684.XA CN111126461B (en) | 2019-12-12 | 2019-12-12 | Intelligent auditing method based on machine learning model interpretation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111126461A CN111126461A (en) | 2020-05-08 |
CN111126461B true CN111126461B (en) | 2023-08-11 |
Family
ID=70499804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911272684.XA Active CN111126461B (en) | 2019-12-12 | 2019-12-12 | Intelligent auditing method based on machine learning model interpretation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111126461B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111612437B (en) * | 2020-06-03 | 2023-09-26 | 云南电网有限责任公司 | Audit operation guiding method and device |
CN112184143B (en) * | 2020-09-07 | 2022-04-29 | 支付宝(杭州)信息技术有限公司 | Model training method, device and equipment in compliance audit rule |
CN112101574B (en) * | 2020-11-20 | 2021-03-02 | 成都数联铭品科技有限公司 | Machine learning supervised model interpretation method, system and equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344941A (en) * | 2008-08-21 | 2009-01-14 | 河北全通通信有限公司 | Intelligent auditing decision tree generation method of 4A management platform |
CN107944999A (en) * | 2017-11-09 | 2018-04-20 | 用友网络科技股份有限公司 | A kind of automatic execution method and device of audit expenses project |
CN110334212A (en) * | 2019-07-01 | 2019-10-15 | 南京审计大学 | A kind of territoriality audit knowledge mapping construction method based on machine learning |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7672884B2 (en) * | 2004-04-07 | 2010-03-02 | Simpliance, Inc. | Method and system for rule-base compliance, certification and risk mitigation |
US8140362B2 (en) * | 2005-08-30 | 2012-03-20 | International Business Machines Corporation | Automatically processing dynamic business rules in a content management system |
US7953688B2 (en) * | 2007-05-08 | 2011-05-31 | Sharon Sadeh | Method and system for facilitating a compliance audit using a rule set |
-
2019
- 2019-12-12 CN CN201911272684.XA patent/CN111126461B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344941A (en) * | 2008-08-21 | 2009-01-14 | 河北全通通信有限公司 | Intelligent auditing decision tree generation method of 4A management platform |
CN107944999A (en) * | 2017-11-09 | 2018-04-20 | 用友网络科技股份有限公司 | A kind of automatic execution method and device of audit expenses project |
CN110334212A (en) * | 2019-07-01 | 2019-10-15 | 南京审计大学 | A kind of territoriality audit knowledge mapping construction method based on machine learning |
Also Published As
Publication number | Publication date |
---|---|
CN111126461A (en) | 2020-05-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111126461B (en) | Intelligent auditing method based on machine learning model interpretation | |
Kharlamov et al. | Ontology based data access in Statoil | |
CN108027818B (en) | Inquiry based on figure | |
US9542445B2 (en) | Using object awareness for a join pruning mechanism | |
US20160055233A1 (en) | Pre-join tags for entity-relationship modeling of databases | |
CN109299100B (en) | Managing internal memory data and the method and system for safeguarding data in memory | |
US7836037B2 (en) | Selection of rows and values from indexes with updates | |
CN106547809A (en) | Complex relation is represented in chart database | |
US20170255709A1 (en) | Atomic updating of graph database index structures | |
CN110472068A (en) | Big data processing method, equipment and medium based on heterogeneous distributed knowledge mapping | |
US20170255708A1 (en) | Index structures for graph databases | |
CN113191497B (en) | Knowledge graph construction method and system for substation site selection | |
Khan et al. | Data tweening: incremental visualization of data transforms | |
US10089343B2 (en) | Automated analysis of data reports to determine data structure and to perform automated data processing | |
US10445370B2 (en) | Compound indexes for graph databases | |
US20160124932A1 (en) | Data processing device and method | |
CN113157943A (en) | Distributed storage and visual query processing method for large-scale financial knowledge map | |
CN115203337A (en) | Database metadata relation knowledge graph generation method | |
CN112069817A (en) | Student knowledge extraction and fusion method and device | |
Di Tria et al. | Research data mart in an academic system | |
Rabuzin | Deductive data warehouses | |
Negi et al. | Unshackling Database Benchmarking from Synthetic Workloads | |
CN117933206B (en) | Service data processing method, device, equipment, storage medium and program product | |
CN113590610B (en) | Blood relationship expression method based on Elastic Search | |
Ravi et al. | A Study on Comparison of UML and ER Diagram |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: 030032 Room 0402, Floor 4, Unit 6, Building 1, No. 79, Tanghuai Road, Tanghuai Park, Comprehensive Reform Demonstration Zone, Taiyuan City, Shanxi Province Applicant after: Lanyun Technology Co.,Ltd. Address before: 030032 Room 0402, Floor 4, Unit 6, Building 1, No. 79, Tanghuai Road, Tanghuai Park, Comprehensive Reform Demonstration Zone, Taiyuan City, Shanxi Province Applicant before: Shanxi Yunhai Zhihua Big Data Technology Co.,Ltd. |
|
CB02 | Change of applicant information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |