Study on the Financial Audit Risk Prevention based on Data Mining Technology

开篇：润墨网以专业的文秘视角，为您筛选了一篇Study on the Financial Audit Risk Prevention based on Data Mining Technology范文，如需获取更多写作素材，在线客服老师一对一协助。欢迎您的阅读与分享！

Abstract. Along with the development of the modern economy， the financial industry is becoming more and more important. However， the global economy has begun to decline after the financial crisis occurred， so that the financial industry is seriously hit， the financial audit risks are greatly increased， and also the security problems of financial audit have gradually become the focus in the eyes of people. In this paper， the risks in the current financial audit and the necessary measures of preventing these risks through data mining technology are analyzed in depth from the perspective of data mining technology as advanced audit technology.

Keywords： Data Mining； Financial Audit； Risk Prevention

Introduction

Along with the rapid development of today's network information technology， how to effectively process vast amounts of information has become the major problem necessary for people to solve， while it is solved by people using computer data mining technology. Also， data mining technology is beginning to play an important role in the financial audit， and also provides a scientific and effective method for processing a large amount of data generated in the financial audit. However， the role of data mining technology in the prevention of the financial audit risks should be never ignored.

An overview of data mining technology

The intension of data mining

A common definition to data mining is as follows： it is a process of extracting the implicit， but potential and useful information and knowledge from the large， incomplete， noisy， fuzzy， random， and practical application data， in which the information and knowledge are unknown by people in advance. For data mining technology， it is unnecessary to create not only any index and also a clear construction for mining the implicit unknown information.

The methods of data mining technology

Data mining technology includes genetic algorithm， clustering analysis， rule induction， neural network， decision tree， decision tree analysis， outlier analysis， etc.

First， genetic algorithm is mainly to find the common features and rules from the object-oriented database. The algorithm is mainly based on the theory of biological natural selection and heredity.

Second， clustering analysis is to analyze the multiple classes of the abstract objects （the objects in each class are similar）. In this way， a global distribution model can be more quickly found， and unimportant information can be eliminated and useful information is preserved.

Third， rule induction is mainly used in the field of knowledge discovery， and can be divided into generalized association rules and quantitative association rules.

Fourth， neural network possesses good self-organizing and adaptability， and is mainly used for processing the highly nonlinear data with the variables of an interactive effect. It is divided into the feed-forward neural network model， the feedback neural network model， and the self-organizing mapping method.

Fifth， decision tree is to display all objects using a form of tree and then classify them， and then respectively seek the useful information. In this method， CART and CHAID are commonly used. This method is fast and features simple classification， so it is applicable to processing the large-scale data.

Sixth， outlier analysis is used for seeking the information different obviously from other objects in the entire data. With this method， the abnormal values can be quickly eliminated from the entire data.

The causes of financial audit risks

Financial audit risks refer to the false audit decisions made by the auditing personnel because the false behaviors in financial enterprise’s statement of assets and liabilities and other discipline-violation problems are not timely discovered and disclosed. Financial audit risks exist in each audit procedure， and the causes of these financial audit risks are mainly concluded as follows：

Weak awareness of auditing personnel

Insufficient attention paid to the auditing personnel to ideological concepts and awareness is an important cause of financial audit risks to occur.

Unqualified quality of auditing personnel

At present， most auditing personnel do not possess professional financial knowledge and technical title， or do not have rich experiment in auditing work in spite of professional knowledge and education. Their ability to judge the impact of the financial operations and comprehensively solve problems is insufficient， so that they can’t completely adapt to the financial audit work.

The absence of quality control standards

At present， specific contents and standards are absent in China’s related audit quality control standards， and also there are no scientific contents and methods provided for the normative steps and methods of the judgment foundation and audit work at some important levels， and for the evaluation standards of audit risks.

The absence of scientific audit management

Some rules and systems have been established in China’s audit standards， but these systems and specifications are not strictly followed by the auditing personnel in fact. They tend to randomly do audit work， but not to develop specific audit elements， so that the audit work reports are written incompletely and irregularly， and even are inconsistent with the audit work papers， etc.

Backward audit technology

At present， the traditional audit methods cannot meet the needs of the current audit work. With the development of information technology， financial audit has begun to gradually develop toward computer audit. However， the development speed is slow， and the increasing speed of data is not followed.

The current situation of the data mining technology study in the field of financial audit

The study of data mining technology was started relatively late in China， but also some good results have been also made along the development of the recent years.

Audit data analysis

The application of data mining technology to financial audit is specifically analyzed in HU Rong and CHEN Yuekun’s paper Data Mining―A New Method of the Modern Audit Data Process written in 2004. The advantages of data mining technology are showed in LV Xinmin and WANG Xuerong’s paper Study in the Application of Data Mining to Audit Data Analysis written in 2007， and also a powerful demonstration for the application of data mining technology to audit data analysis is proposed.

The audit process

The work flow of data mining technology in the whole audit process is early expounded in YI Renping’s paper the Audit Model Framework based on Data Mining written in 2003.

Audit risk decision-making

In CHEN Danping’s paper Study on the Audit Risk Decision-making in the Data Mining Model written in 2007， the study of audit risks using data mining technology is proposed， and also the theory of reducing audit risks using data mining technology is proposed.

The financial audit process based on data mining technology

The acquisition of audit data

According to audit contents and purposes， the collected original audit data is required to import into the database of the audit system.

Data preprocessing

The data is cleaned and classified before data mining， so as to get rid of useless information and noise removing the irrelevant data.

The analysis of data mining

After the first two steps， a good preparation has been made for data mining. According to the actual situation， the mining analysis of the database is conducted by choosing the most appropriate data mining technology.

The audit processing

After an analysis on the data mining of the database， the data generated in the analysis of data mining can be analyzed and processed by auditors and experts， and then is added into the knowledge base.

The newly-added audit data processing

In processing the newly-added audit data， the data is necessarily pre-processed using the above-mentioned data preprocessing method， and then the data audit analysis is conducted using the updated knowledge base as the judgment standards. If suspicious data is found， it is analyzed and processed by auditors and experts， and then is added into the knowledge base.

The application of data mining technology to the prevention of financial audit risks

Substantive procedure

With the help of data mining technology， auditing personnel can quickly screen out the data samples， so that the efficiency of audit work is not only improved， and also the emergence of audit risks is reduced.

In conducting the financial audit of loan enterprises’ management conditions， the enterprises with the similar financial indexes can be classified into a group using the clustering analysis method， so that those enterprises which can repay the loan on schedule can be determined， those necessary to develop the repayment of loan can be positioned， and those unable to repay the loan on schedule can be known.

In auditing the substantiality of enterprise’s interest receivable， the accounting data with the similar indexes is classified into a group using the clustering analysis method. Thus， the interest receivable obviously different from those of other months， the interest receivable which is repeated in an account， and the same interest receivable which is not equal to the amount in general ledger and detail account can be found clearly. In auditing the financial affairs of financial enterprises， some financial indicators can be statistically classified and analyzed using the statistical analysis method.

In auditing the economic data of enterprises， a certain relationship between different types of data can be quickly found using the rule induction method， and simultaneously the relationship among the income statement， the cash flow statement， and the balance sheet analysis can be analyzed and searched according to the law different from the financial logic relationship， so as to help auditing personnel find out some unknown economic activities.

Testing the internal system

When internal control is tested using the decision tree analysis method of data mining technology， the data necessary to audit in the financial enterprises is classified， and then part of the data is chosen from each type of data for the testing. In testing the internal control， there is a problem in the internal control system if isolated data can be found by the auditing personnel.

Evaluating financial enterprises

To evaluate the risk of the enterprises， the previous financial statements are necessarily extracted from the mass data in the database of the financial enterprises to analyze using the data generalization method of data mining technology， so as to determine the nature of a financial enterprises and the condition of this industry.

To judge whether there is a false financial report in some financial enterprises， the mutual cooperation of multiple data mining techniques is necessary. It is necessary to mine the mass data in the database of the financial enterprise using data preprocessing technology， and also specifically express the features of the financial data using multiple methods such as mean value method， hierarchical method， and statistical method， so that auditing personnel are easy to make analysis and judgment.

Conclusion

In the financial audit work， the application of computer’s data mining technology to processing and extracting the mass data generated in the audit work plays an irreplaceable role in increasing the efficiency of the financial audit work and preventing the risks of financial audit.

References

[1] CHEN Guanping. Study on the Ways for Implementing the Functions of the Financial Audit Immune System―An Empirical Analysis based on the Audit Results [J]. Journal of Auditing Research， 2009.3.

[2] LIU Yanjiao， CHANG Jingjing. The Design and Implementation of the Intelligent Audit System based on data mining [J]. Journal of Financial Audit， 2011.6.

[3] CHENG Guanhua. Study on the Application of Data Mining technology in the Commercial Banks Internal Auditing [J]. Journal of New Accounting， 2011.3.

[4] LU Jiahui， WEN Huayi. Study on the Computer Audit Model of Commercial Banks [D]. Beijing： China times economic publishing house， 2009.

[5] CHEN Yuekun. data mining―New Data Processing Methods for the Modern Audit [J]. Chinese audit. 2011.7

[6] LV Xinmin， WANG Xuerong. Study on the Application of Data Mining Technology in the Audit Data Analysis [J]. Audit and Economic Research. 2013.3.