臺灣博碩士論文加值系統

本研究提出一個以自組織映射網路執行群集運算的資料探勘法。本法使用資訊樣型為基的資訊採擷法解析群集後資料內部的特徵，發掘潛藏的有用資訊。研究目標係使自組織映射網路能有效地應用於資料庫內的知識發掘。本研究提出的資訊樣型比對法，係進行群集後資料的交叉比對分析。透過反覆逐步擴充的搜尋運算，將資料內部的知識以資訊樣型方式表達出來。此外為使資訊的表達具有口語性，整個探勘法設計了一個資訊語意對映機制，把各群各屬性的資料分布轉換成相對的歸屬等級。歸屬等級是以符合資料型態的口語語詞來表示，使得資訊樣型所呈現的內容更易於解讀。本文並發展一種強化群集的方法，定義出一個資料框聚係數，藉著改變此係數來達到強化群集的效果。過程是逐步適度的將群集的標準放寬，避免損失資料中隱含的有益資訊。本研究並以實例來驗證所提的架構與流程，驗證結果顯示本法能有效地採擷資料的特徵，較其它方法更能擷取不完全資訊及克服雜訊。

This thesis presents a data mining method that uses Self-Organizing Map to execute data clustering. Information extracting techniques for pattern search are used to analyze clustered data and to explore useful information. The goal of this research is to postulate the Self-Organizing Map for knowledge discovering in databases. The thesis defines several generic information patterns and data characteristics of the clustered data. This research presents a data coherence enhancement algorithm to intensify the resulting data groups. By gradually relaxing the data grouping restrictions, which is imposed by a data framing factor, data are grouped iteratively without losing useful information. To make information expression be more colloquial, a mapping mechanism is developed to change the data with various attributes into the ones that are easier comprehended. Ranks regarding to the data values are thus defined with an oral expression. Examples are tested to verify the presented data mining procedures. Results show that the method can effectively extract features of the clustered data. Results also show that the presented method is capable of extracting incomplete information and overcoming noise.

謝誌...........................................i
摘要..........................................ii
Abstract.....................................iii
目錄..........................................iv
圖目錄.......................................vii
表目錄......................................viii
名詞彙編.......................................x
符號列表....................................xiii
第一章 緒論...................................1
1.1 研究背景及動機.............................1
1.2 研究目的...................................2
1.3 研究範疇...................................3
1.4 研究方法及架構.............................4
1.5 章節概要...................................5
第二章 文獻探討...............................7
2.1 資料庫內的知識發掘.........................7
2.1.1 資料庫內的知識發掘的意義.................7
2.1.2 資料庫內的知識發掘的程序.................8
2.2 資料探勘..................................10
2.2.1 資料探勘的方法..........................10
2.2.2 資料探勘的技術..........................11
2.2.3 資料探勘的應用..........................13
第三章 類神經網路技術........................16
3.1 基本理論..................................16
3.1.1 基本名詞................................16
3.1.2 基本架構................................19
3.1.3 目前發展及應用..........................21
3.2 自組織映射網路............................21
3.2.1 特有名詞................................22
3.2.2 網路架構................................25
3.2.3 網路演算法..............................27
3.2.4 演算釋例................................29
第四章 以自組織映射網路為基的資料探勘流程.....33
4.1 資料探勘流程..............................33
4.2 資料前處理................................36
4.2.1 資料格式................................36
4.2.2 資料前處理流程..........................37
4.2.3 資料前處理釋例..........................38
4.3 資料群集劃分法則..........................40
4.3.1 分群門檻為基的群集劃分演算法............40
4.3.2 群集劃分釋例............................42
4.4 邊緣資料刪除法則..........................44
4.4.1 資料框聚係數............................45
4.4.2 群集強化演算法..........................46
4.4.3 群集強化釋例............................47
4.5 資料特徵表現法則..........................49
4.5.1 資料群集特徵值計算法....................49
4.5.2 特徵計算釋例............................50
4.6 歸屬等級對映法則..........................52
4.6.1 口語語詞表示法..........................52
4.6.2 歸屬等級對映演算法......................53
4.6.3 歸屬等級對映釋例........................55
4.7 資訊樣型比對法則..........................58
4.7.1 資訊樣型定義............................59
4.7.2 資訊樣型比對演算法......................62
4.7.3 資訊樣型比對釋例........................64
第五章 實例驗證及探討........................73
5.1 驗證資料..................................73
5.2 實例資料..................................88
第六章 結論與未來展望........................94
6.1 總結......................................94
6.2 未來展望..................................94
參考文獻......................................96
附錄..........................................99

謝俊宏，植基於類神經網路的分類模式建構之研究-以財務資料應用為
例，博士論文，私立淡江大學管理科學研究所，1998。
Andrews, R., Diederich, J., and Tickle, A.B., “Survey and
Critique of Techniques for Extracting Rules from Trained
Artificial Neural Networks,” Knowledge-Based Systems, Vol.
8, No. 6, pp. 373-389, 1995.
Agrawal, R., Imielinski, T., and Swami, A., “Database Mining:
A Performance Perspective,” IEEE Transactions on Knowledge
and Data Engineering, Vol. 5, No. 6, pp. 914-925, 1993.
Chatterjee, C. and Roychowdhury, P., “On Self-Organizing
Algorithms and Networks for Class-Separability Features,”
IEEE Transactions on Neural Networks, Vol. 8, No. 3, pp. 663-
678, 1997.
Chen, M.S., Han, J., and Yu, S., “Database Mining: A Overview
from a Database Perspective,” IEEE Transactions on Knowledge
and Data Engineering, Vol. 8, No. 6, pp. 866-883, 1996.
Fayyad, U., “Data Mining and Knowledge Discovery in Databases:
Implication for Scientific Databases,” Scientific and
Statistical Database Management, pp. 2-11, 1997.
Fayyad, U., Haussler, D., and Stolorz, P., “Mining Scientific
Data,” Communications of the ACM, Vol. 39, No. 11, pp. 51-
57, 1996.
Ferguson, C.J., Lees, B., MacArthur, E., and Irgens C., “An
Application of Data Mining for Product Design,” IEE
Colloquium on Knowledge Discovery and Data Mining, 1998.
Imielinski, T. and Mannila, H., “A Database Perspective on
Knowledge Discovery,” Communications of the ACM, Vol. 39,
No. 11, pp. 58-64, 1996.
Joutsensalo, J. and Miettinen, A., “Self-Organizing Operator
Map for Nonlinear Dimension Reduction,” Proceedings of
International Conference on Neural Networks (ICNN''95), Vol.
1, pp. 111-114, 1995.
Kohonen, T., “The Self-Organizing Map,” Proceedings of IEEE,
Vol. 78, pp. 1464-1480, 1990.
Kohonen, T., “Generalizations of The Self-Organizing Map,”
Proceedings of International Joint Conference on Neural
Networks (IJCNN''93), Vol. 1, pp. 457-462, 1993.
Kohonen, T., “Things You Haven''t Heard about the Self-
Organizing Map,” IEEE International Conference on Neural
Networks, Vol. 3, pp. 1147-1156, 1993.
Kohonen, T., “New Developments and Applications of Self-
Organizing Maps,” Proceedings of International workshop on
1996, pp. 164-172, 1996.
Lu, H., Setiono, R., and Lin, H., “Effective Data Mining Using
Neural Networks,” IEEE Transactions on Knowledge and Data
Engineering, Vol. 8, No. 6, pp. 957-961, 1996.
Nair, G. J. and Narendran , T. T., “Case: A clustering
algorithm for cell formation with sequence data,”
International Journal of Production Research, Vol.36, No. 1,
pp. 157-179, 1998.
Pham, D. T. and Bayro-Corrochano, E. J., “Self-Organizing
Neural-Network-Based Pattern Clustering Method with Fuzzy
Outputs,” Pattern Recognition, Vol. 27, No. 8, pp. 1103-
1110, 1994.
Piatetsky-Shapiro, G., “The Data Mining Industry Coming of
Age,” IEEE Transactions on Intelligent Systems, Vol. 14, pp.
32-34, 1999.
Samad, T. and Harp, S.A., “Feature Map Learning with Partial
Training Data,” Proceedings of International Joint
Conference on Neural Networks (IJCNN''91), Vol. 2, pp. 949,
1991.
Santosa, F., “Data Mining: An Industrial Research
Perspective,” IEEE Transactions on Computational Science and
Engineeaing, Vol. 42, pp. 6-9, 1997.
Sarle, W.S., “Neural Networks and Statistical Models,”
Proceedings of the Nineteenth Annual SAS Users Group
International Conference, April, 1994.
Vesanto, J. and Alhoniemi, E., “Clustering of the Self-
Organizing Map,” IEEE Transactions on Neural Networks, Vol.
11, No. 3, pp. 586-600, 2000.
Zhang, X. and Li, Y., “Self-Organizing Map as a New Method for
Clustering and Data Analysis,” Proceedings of International
Joint Conference on Neural Networks (IJCNN''93), pp. 2448-
2451, 1993.