Dissertation > Excellent graduate degree dissertation topics show

Research on Method of Data Mining Based on Granular Computing Model

Author: ZhouJun
Tutor: LinQing
School: Jiangsu University
Course: Applied Computer Technology
Keywords: data mining rough set quotient granularity dynamic granularity classification
CLC: TP311.13
Type: Master's thesis
Year: 2010
Downloads: 243
Quote: 2
Read: Download Dissertation

Abstract


Data mining, is a nontrivial process of acquiring the effective, new, potential useful, and ultimately comprehensible pattern from the mass data. The classification, having the wide application prospect in the commercial domain, is an important kind of data mining question. At first training set classified should be selected from the data, and then data mining classification technology is used to establish the classification models on this training set; classify the rest without classification.Granular Computing’s thought is produced in 1970s, whose basic thought is imitating the human’s way of thinking: when ponder and solve the identical question, or observe it from overall to partition, collect and analyze various aspects of the situation; Contrarily, collect data from part to whole; or find out the different perspective problems from various angles , finally Comprehensive analysis. In recent years, Granular Computing starts to be applied into the data mining, which has made some achievements and became a new research direction in data mining at present.This thesis summarized the principles and actuality of the data mining at first, from the data mining and the knowledge classification angle, take the rough collection theory as the foundation, using for reference some known theory fruits of soft computing, the thesis lucubrate granular computing. The major contents can be separated into some parts listed below:(1) Studying the classification problem of information system based on the decision tree model. Based on Rough Set, combined with the principle that knowledge relation has the granularity nature, the degree of association between condition attribute set and decision kind set selects decision kind set, so that is the quotient granularity. Take the decision tree method as the theoretical basis, based on knowledge rough’s granularity principle, the quotient granularity being applied into the decision tree method, a new method to design the decision tree is proposed. The advantages of this method are described in detail. Case study has proved that the way of quotient granularity construction algorithm proposed for the decision tree is reliable and effective.(2) Studying the attribute reduction based on the dynamic granularity theory. Rough set theory is an effective approach to imprecision, vagueness, and incompleteness in classification analysis and know-ledge discovery. Rough set based on boundary region is dynamic analyzed from the coarser degree of granularity. According to dynamic quotient granularity’s definition from the principle of attribute connection, a new attribute reduction algorithm based on the coarser degree of granularity principle is proposed. The optimal reduction set can be selected from all reduction set with algorithm for dynamic quotient granularity. It abandons the tradition to ask the core first, then chose the optimal reduction set. The validity of proposed granularity computing algorithm is proved by the application of practical database.(3) Studying the system classification question of the incomplete information. According to human’s cognition rule, that is, the humanity may use the limited knowledge, to gain the quite satisfactory result in the coarse granularity level, to avoid the incomplete characteristic in the fine granularity level. Based on Granular Computing model, the union theory of Quotient Space and Rough Set theory, the projection and coarse granularity processing to the default training sample data set, enables the system processed to become the policy-making definite system. It’s possible to use the existing known sample , use the multi-granularity level processing method at multi-level of granularity to solve the classification problem in the incomplete information system, and overcome the classification problem that the majority algorithm is only able to be applied in the complete information system.Main conclusions and significances of this thesis are as following:(1) The quotient granularity is defined from the attribute connection, applies in the data class modelling.(2) From the dynamic quotient granularity angle, the optimal reduction set can be selected from all reduction set with algorithm for dynamic quotient granularity.(3) Based on Granular Computing model, the union business space and Rough the Set theory applies in the solution incomplete information system classified question.At last, it establishes a system for data mining in management of teaching quality, applying the algorithms put forward in the thesis, for attribution reduction and optimum rule extraction to fulfill the functions of the model.

Related Dissertations

  1. K Company’s Improving Planning and Forecasting for the Reasonable Allocation of Inventory,F224
  2. The Simulation and Analysis of Spectrum Sensing Based on Higher Order Cumulants,TN911.23
  3. Research on the Classification Based on the Reconstruction of Solder Joint,TP391.41
  4. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  5. Tourism Comments on the Internet’s Semantic Analysis and Usefulness Research,TP391.1
  6. Fault Diagnosis Method Based on Support Vector Machine,TP18
  7. Uptake and Accumulation of Heavy Metals of Pistia Stratiotes and Eicharnia Crassipes under Combinated Pollution Condition,X173
  8. A Study on Healthcare Product Marketing Based on Data Mining Technology,F426.72
  9. Gao Zhong-ying academic thought and experience and use of Bufei Decoction treatment of common diseases of the respiratory system drug law,R249.2
  10. Bing- thick academic thought and clinical experience and empirical studies apply to turtle soups treatment of chronic kidney disease,R249.2
  11. Research on Nondestructive Detection Technology for External Qualities of Papayas Based-on Vision,S667.9
  12. Comparison of Gene Expression Data Cluster Methods and Gene Network Construction for Phytophthora Sojae Genes,S435.651
  13. Study on Growth Monitoring Technique Based on Pixel Un-Mixing Method and HJ Remote Sensing Images in Paddy Rice,S511
  14. The Design and Implementation of Bicluster Data Analyzing Software,TP311.52
  15. Research on Clustering Algorithm Based on Mutation Particle Swarm Optimization,TP18
  16. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  17. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  18. Based on Rough Set of Urban Areas When Traffic Green Control System Research,TP18
  19. Based on data mining research tax audit case selection,F812.42
  20. Packaging waste processed Legal Issues,D922.6
  21. Community-oriented education, personalized learning system and its implementation,TP391.6

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net  Mobile