Dissertation > Excellent graduate degree dissertation topics show

The Research on Key Techniques of Financial Fraud Identification Based on Dynamic Combination of Classifiers

Author: YuHaiPing
Tutor: ZhuYuQuan；ChenGeng
School: Jiangsu University
Course: Applied Computer Technology
Keywords: The identification of financial fraud Clustering is divided Multi - classifier dynamic combination Decision Tree Rough Set Feature Selection
CLC: F275
Type: Master's thesis
Year: 2010
Downloads: 119
Quote: 1
Read: Download Dissertation

Abstract


In the modern market economy, the company's financial bullying Jo behavior in an endless stream, had a huge impact on the securities market, triggering an unprecedented credit crisis. Therefore, to identify financial fraud is particularly important. Classification techniques in data mining, the company's financial data analysis, processing, data mining, which contains information and rules to help investors and accountants to easily cope with a variety of complex financial data behavior, has a high academic value and broad application prospects. Currently, the classification of financial fraud identification technology research has just started, existing classification methods applied directly to the identification of financial fraud are still many problems. Exploration and research for the classification of financial fraud identification has a very important practical significance. The paper describes the financial fraud for the purpose of identification, significance and status quo and existing problems in the classification of financial fraud data of their own characteristics and existing portfolio, proposed a dynamic combination of classification method based on clustering by use of the company's financial The data on the effectiveness of the method was validated, and financial fraud detection prototype system is designed and implemented using object-oriented technology. The main work of this paper is as follows: 1, describes the combination of dynamic classification and identification of fraud, Research, discussed in detail based on the classification of financial fraud to identify the basic steps, and describes the most commonly used classification method. The proposed based rough set theory Decision Tree S_D_Tree of. The method uses the attribute importance in rough set theory instead of the traditional methods of information gain ratio as the standard of the selection of the test attribute. At the same time, introduced in the process of constructing a decision tree Failnode-prune pruning strategy to achieve the purpose of the simplified decision tree. 3, based on clustering by the dynamic combination classification DCC-CD. First, according to the data class distribution imbalance characteristics, the use of PAM clustering algorithm to divide restructuring, S_D_Tree approach members were trained classifier fusion results, and finally through the dynamic combination of classification output. DCC-CD method in recognition of the company's financial fraud. Selection method based on the characteristics of the genetic search the best subset of attributes, and the practical application of DCC-CD method classification performance has been verified. At the same time, the use of object-oriented technology to design and implement a financial fraud detection system.

Related Dissertations

  1. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  2. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  3. Fault Diagnosis Method Based on Support Vector Machine,TP18
  4. Research on Feature Selection and Construction in Emotion Speech Recognition,TP18
  5. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  6. Based on Rough Set of Urban Areas When Traffic Green Control System Research,TP18
  7. Design and Development of Teaching Quality Assessment System Based on Data Mining,TP311.13
  8. Based on Data Distribution Characteristics of Text Classification,TP391.1
  9. Incremental rough set attribute reduction,TP18
  10. Calculation of Knowledge Granulation and Study of Its Application in Attribute Reduction,TP18
  11. Research of License Plate Recognition Based on Rough Sets and Fuzzy SVM,TP391.41
  12. Research and Implementation of a Dynamic Feature Selection Method for Vehicle Recognition System,TP391.41
  13. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  14. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  15. Application of Rough Set and Flex in Mid-long Term Runoff Forecasting,P338
  16. Importance of the study of the physical and chemical indicators based on rough set theory Daqu,TS262.3
  17. The software design and implementation of the the clothing quality prediction system,TP311.52
  18. Water quality time series data processing and Early Warning System Construction Research Database,TP274
  19. Study on the Decision Tree Classification Algorithm and Its Application Based on Rough Set Theory,TP18
  20. Based on the combined effect of the rough planning model,O221
  21. Based on the core set of examples of attribute reduction method,O159

CLC: > Economic > Economic planning and management > Enterprise economy > Corporate Financial Management
© 2012 www.DissertationTopic.Net  Mobile