臺灣博碩士論文加值系統

自動化生產過程有上百個步驟，每個步驟都包含相當多的測量項目，所以會得到相當龐大的原始數據。這些數據不但有極大量的變數，同時變數之間也存在高度的關聯性，因此會有多餘資訊的產生。依據這些特性我們選擇使用圖形化高斯模型的方法建立模型，提供研究者分析資料中解釋變數對於被解釋變數的影響以及解釋變數之間互相的關聯性。
以晶圓製造為例，論文中將資料作前置處理後，先就一般的方法作討論，並測試其計算的極限和一些影響的因素。接著結合因素分析和多維縮放比例的方法進行變數的分群，藉由分群建立模型的方式來簡化一般方法，並且針對偏差值提出改進。利用簡化方法所建立的模型在分析更多變數量的同時，更能使模型偏差值保持於使用者要求的範圍之內。
同樣的方法亦可應用於與製程類似自動化過程所測量的大量數值資料中，模型建立之後再結合專家系統或貝氏網路，將可以對結果進行預測與診斷之工作。

There are hundreds of steps in the process of automated manufacture operation. Every step contains lots of measurements. As a result a tremendous amount of data is available. These data have a great deal of variables, which are highly correlated. According redundancy exits. In order to provide analysts the influence of predictors upon dependents, and to explain the correlations of variables, we use Graphical Gaussian Models(GGMs) to establish models based on the characteristic of the gathered data.
Take manufacture of silicon wafers for example, data will be preprocessed first. Then we will discuss the measured limits and the factors of the general GGMs. Through the combination between factor analysis (FA)and multidimensional scaling (MDS), clusters of variables will be proceeded. According to the clusters, he procedure of modeling will be simplified and an improved method will be introduced to analyze more variables while maintaining the requested deviance.
This method also can be applied to the massive data gathered by the similar procedure like automated manufacturing operation. Combining Expert system or Bayesian Network, we can prognosis and diagnosis results after a model is built.

中文摘要 i
英文摘要 ii
誌 謝 iii
目 錄 iv
表 目 錄 vii
圖 目 錄 viii
第一章 緒論 1
1.1 研究動機 1
1.2 研究方法 3
1.3 論文結構 7
第二章 前置處理和多變量分析 8
2.1 前置處理 8
2.1.1 前置處理簡介 8
2.1.2 歧異值的處理 8
2.1.3 缺少資料項的處理 10
2.1.4 特徵選取 10
A. 使用特徵選取的原因 10
B. 特徵選取的方法 11
2.2 多變量分析 13
2.2.1 多變量分析簡介 13
2.2.2 因素分析 13
2.2.3 多維縮放比例 16
第三章 圖形化模型 19
3.1 圖形化模型簡介 19
3.2 圖形化高斯模型原理 20
3.2.1 圖學簡介 20
3.2.2 圖形化模型的理論根據 23
3.2.3 最大近似估測 25
3.2.4 最大基數搜尋測試 28
3.2.5 測試和偏差值 30
3.3 圖形化高斯模型建立流程 31
第四章 簡化方法的問題和改進 34
4.1 最大近似估計的問題 34
4.2 降低簡化方法的偏差值 34
4.3 簡化方法的完整步驟 35
第五章 實驗與討論 38
5.1 實驗資料簡介 38
5.2 前置處理 39
5.2.1 歧異值的問題 39
A. 找尋歧異值 39
B. 歧異值對特徵選取的影響 39
C. 歧異值對模型建立的影響 40
5.2.2 缺少資料項的問題 42
5.3 GGMS效能檢測 42
5.3.1 變數個數對計算時間的影響 42
5.3.2 不同檢測值對模型的影響 44
5.3.3 GGMs過程中模型的變化 48
5.4 簡化方法 49
5.4.1 資料分群 49
5.4.2 簡化GGMs的建立 53
A. 分群建立模型 53
B. 簡化方法的改進 54
第六章 結論 58
附錄A F-測試 60
參考文獻 62

[1] M. S. Grewal and A. P. Andrews., Kalman Filtering: Theory and Practice, Prentice-Hall, 1993.
[2] B. Frey., Graphical Models for Machine Learning and Digital Communication, MIT Press, 1998.
[3] David Edwards, Introduction to Graphical Modelling, Second Edition, Springer- Verlag, 2000.
[4] J. Whittaker, Graphical Models in Applied Multivariate Statistics, John Wiley, New York, 1990.
[5] T. W. Anderson, An Introduction to Multivariate Statistical Analysis, Second Edition, John Wiley, New York, 1984.
[6] Henry Stark, John W. Woods, Probability and Random Processes with Applications to Signal Processing, Third Edition, Prentice-Hall, 2002.
[7] Barbara G. Tabachnick, Linda S. Fidell, Using Multivariate Statistics, Third Edition, Harper Collins, New York, 1996.
[8] Castillo E., Gutiérrez J. M., Ali S. Hadi, Expert Systems and Probabilistic Network Models, Springer-Verlag, 1997.
[9] Schoichiro Nakamura, Applied Numerical Methods in C, Prentice-Hall, 1995.
[10] Robert G. Cowell, et al., Probabilistic Networks and Expert Systems, Springer-Verlag, 1999.
[11] Borg I., Gorenen P., Modern Multidimensional Scaling : Theory and Applications, Springer-Verlag, 1997.
[12] 曹永誠，「自動化經驗談」，工業電機&自動控制裝置與設計雜誌，第十五期，全國工業電機自動控制雜誌社，1998年6月

[13] Jeff A. Bilmes, “Graphical Models and Automatic Speech Recognition” , Mathematical Foundations of Speech and Language Processing, Institute of Mathematical Analysis Volumes in Mathematics Series, Springer-Verlag, 2003.
[14] Jeff A. Bilmes, et al., “Discriminatively Structured Graphical Models for Speech”, UWEE Technical Report Series, Summer 2001.
[15] Mark S. “Multidimensional Scaling”, Encyclopedia of Cognitive Science, pp. 1-7, Macmillan Reference Ltd., March 2001.
[16] Man-Suk Oh, Adrian E. Raftery, “Bayesian Multidimensional Scaling and Choice of Dimension”, Journal of the American Statistical Association, 455, 96, pp. 1031-1044, American Statistical Association, September 2001.
[17] Isabelle Guyon, André Elisseeff, “An Introduction to Variable and Feature Selection”, Journal of Machine Learning Research, 3, pp. 1157-1182, MIT Press, March 2003.
[18] Anne Berry, Jean R. S. Blair, Pinar Heggernes, “Maximum Cardinality Search for Computing Minimal Triangulations for Graph”, Algorithmica, 4, 39, pp. 287-298, Springer-Verlag, New York, May 2004.
[19] Kruskal J. B., Wish M., “Multidimensional Scaling”, Sage University Paper series on Quantitative Applications in the Social Sciences, pp. 7-11. Sage Publications, Newbury Park, CA, 1978.
[20] Cattell R. B., “The scree test for the number of factors”, Multivariate Behavioral Research, 1, pp. 245-276, 1966.
[21] J. Makhoul, “Linear prediction: A tutorial review”, Proceedings of the IEEE, 63, 4 , pp. 561-580, April 1975.
[22] T. P. Speed, H. T. Kiiveri, “Gaussian markov distributions over finite graphs” Annals of Statistics, 14, 1, pp. 138-150, 1986

[23] Picci G., Pinzoni S., “On Joint Identification and Latent Variable Estimation in Factor Analysis Models”, Proceedings of the IEEE Conference on Decision and Control, 4,pp. 3877-3882 , Lake Buena Vista, FL USA, December 1994.
[24] Baek K., Draper, B. A., “Factor Analysis for Background Supression”, Proceedings of the IEEE Conference on Pattern Recognition, 2, pp. 643-646, August 2001.