臺灣博碩士論文加值系統

在資訊化為導向的現代社會中，企業對資料庫的依賴愈來愈深，當資料庫中的屬性其屬性值為空值時，會造成整個資料庫不能正常的運作，此時企業將遭受重大的損失。在本研究中，我們提出一個有效且系統化的方法來估計空值，其漸進式的程序來改善效能。此方法主要包含三個優點：(1)屬性選擇；(2)考量資料庫特性；(3)依情境進行屬性正規化。本研究包含二階段：(1)資料前處理；(2)建模。首先資料前處理的階段使用漸進式的方法進行估計空值，主要包含非切割式，硬式切割與模糊切割的方式。第二此模式其主要是利用Beta係數/相關係數與切割方式來計算不同屬性之間的影響程度。最後，為了驗證我們提出之方法，採用二個資料庫:(1) 人力資源資料庫，(2) Waugh資料庫。且利用平均絕對誤差率（MAER）來做為評估準則且做為比較之依據，由MAER的結果顯示本研究所提出的方法能有更準確結果。

A database system generally cannot operate properly if it contains null values of attributes in the system. The study proposes efficient and systematical approaches to estimating null values in a relational database, which progressively improve the performance of the null value. The proposed approaches have three advantages: (1) attribute selection; (2) consideration of database pattern and (3) situation based units of attribute measurement (normalized). The approaches include two phases: (1) data-preprocessing, (2) model construction. Firstly, the data-preprocess phrase uses a progressive approach to estimate null value, which including non-partition, hard partition and fuzzy partition approaches. Secondly, the model building phrase utilizes the beta/correlation coefficient and partition approach to calculate the relative influence of different attributes. Two databases are used to verify the proposed approaches: (1) Human resource database, (2) Waugh’s database. Furthermore, this study uses mean of absolute error rate (MAER) as an evaluation criterion for comparison with other methods. It demonstrates that the proposed approaches are superior to existing methods for estimating null values in relational database systems.

中 文 摘 要 i
Abstract. ii
誌 謝 iii
Content iv
Table Index vi
Figure Index x
1. Introduction 1
1.1. Background 1
1.2. Motivation and research objective 2
1.3. Organization of this dissertation 5
2. Preliminary 6
2.1. Fuzzy set theory 6
2.1.1. The arithmetic operations of fuzzy numbers 6
2.1.2. Linguistic variable 9
2.2. Correlation Metric 10
2.2.1. Correlation coefficient 10
2.2.2. Beta coefficient 12
2.2.3. Fuzzy correlation 15
2.2.4. Similarity 17
2.3. Clustering method 17
2.4. Related works for estimating null value 18
3. The Proposed approaches for in estimating null value 20
3.1. Research Framework 20
3.2. Property comparisons of different approaches 25
4. Estimating null value based on the degree of influence 27
4.1. Algorithm for estimating null values 28
4.2. An example 33
5. Estimating null value based on hard partition approach 58
5.1. Algorithm for estimating null values 60
5.2. An example 63
6. Estimating null value based on fuzzy partition approach 73
6.1. Algorithm for estimating null values based on fuzzy partition approach 73
6.2. An example 76
7. Performance comparison and findings 89
7.1. Performance comparison 89
7.2. Findings 96
8. Conclusions 103
Reference 105

Arnold SF (1990) Mathematical Statistics, Prentice-Hall, Englewood Cliffs NJ.
Babad YM, Hoffer JA (1984) Even no data has a value. Communications of the ACM 27(8):748-757.
Bradley PS, Fayyad U (1968) Refining initial points for k-means clustering. Proceedings of the Fifteenth International Conference on Machine Learning: 91-99
Bradley PS, Fayyad U and Reina C (1998) Scaling Clustering Algorithms to Large Databases. The Fourth International Conference on Knowledge Discovery and Data Mining: 27-31.
Chen SM, Chen HH (2000) Estimating null values in the distributed relational databases environments. Cybernetics and Systems: An International Journal 31(8):851-871
Chen SM, Huang CM (2003) Generating weighted fuzzy rules from relational database systems for estimating null values using genetic algorithms. IEEE Transactions on Fuzzy Systems 11(4): 495-506
Chen SM, Hsiao HR (2005) A new method to estimate null values in relational database systems based on automatic clustering techniques. Information Sciences: An International Journal 169:47-69
Chen SM, Lee SW (2003) A new method to generate fuzzy rules from relational database systems for estimating null values. Cybernetics and Systems: An International Journal 34: 33-57
Chen SM, Yeh MS (1997) Generating fuzzy rules from relational database systems for estimating null values. Cybernetics and Systems: An International Journal 28(2):695-723
Chen SM, Yeh MS (2002) A method for generating fuzzy rules from relational database systems for estimating null values. Intelligent Systems: Technology and Applications. edited by C. T. Leondes, CRC Press, FL: Boca Raton, U. S. A., 4: 157-179
Cheng CH, Chang JR and Yeh CH (2006) Entropy-based and Trapezoid Fuzzification-based Fuzzy time series approaches for forecasting IT project cost, Technological Forecasting and Social Change, 73.
Cheng CH, Lin Y. (2002) Evaluating the best main battle tank using fuzzy decision theory with linguistic criteria evaluation. European Journal of Operational Research 142(1):174-18.
Cheng CH, Cheng GW, Wang JW (to be published in mid-2008) Multi-attribute Fuzzy Time Series Method Based on Fuzzy Clustering. Expert Systems With Application 35(1).
Cheng CH, Wang JW (2006) A new approach for estimating null value in relational database. Soft Computing, 10(2):104-114.
Chiang DA, Lin NP (1999) Correlation of fuzzy sets. Fuzzy Sets and Systems 102: 221-226
Codd EF (1979) Extending the Database Relational Model to Capture More Meaning. ACM Transactions on Database Systems 4(4):397-434
Daniel WW, Terrell JC (1995) Business Statistics for management and economics (7th ed.). Houghton Mifflin Company.
Davidson JW, Savic DA, Walters GA (2003) Symbolic and numerical regression: experiments and applications. Information Sciences 150:95–117
Draper NR, Smith H (1998) Applied Regression Analysis, John Wiley and Sons.
Dubois D, Prade H (1980) Fuzzy Sets and Systems: Theory and Applications. New York, London, Toronto.
Eliez S., Blasey CM, Freund LS, Hastie T and Reiss AL (2001) Brain anatomy, gender and IQ in children and adolescents with fragile X syndrome. Brain 124: 1610-1618.
Forgy E (1965) Cluster analysis of multivariate data: Efficiency vs. interpretability of classifications. Biometrics 21:768.
Gate Comm Software (2005) http://www.basenow.com/help/Null_values.asp
Hair JF Jr., Anderson RE, Tatham RL, Black BW (1998) Multivariate Data Analysis With Readings. Fourth Edition, Prentice-Hall International Co.
Han J, Kamber M (2000) Data Mining: Concepts and Techniques. Morgan Kaufmann, New York
Hong TP, Lee CY (1996) Induction of fuzzy rules and membership functions from training examples. Fuzzy Sets and Systems, 84(1): 33-47.
Hsieh CH, Chen SH (1999) Similarity of generalized fuzzy numbers with graded mean integration representation. in Proc. 8th Int. Fuzzy Systems Association World Congr. 2:551-555
Huang X, Zhu Q (2001) A pseudo-nearest-neighbor approach for missing data recovery on Gaussian random data sets. Pattern Recognition Letter 23:1613-1622
Kaufmann A, Gupta MM (1985) Introduction to Fuzzy Arithmetic, Van Nostrand, New York.
Konstam A (1998) Group classification using a mix of genetic programming and genetic algorithms, ACM Press.
Mangasarian OL, Wolberg WH (1990) Cancer diagnosis via linear programming, SIAM News 23:1-18
Mangasarian OL, Setiono R. and Wolberg WH (1990) Pattern recognition via linear programming: Theory and application to medical diagnosis, in: T.F. Coleman, Y. Li (Eds.), Large-scale numerical optimization, SIAM Publications, Philadelphia :22±30.
MacQueen JB (1967) Some methods for classification and analysis of multivariate observations. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability 1:281-297
McClave TJ, Benson GP, Sincich T (2001) Statistics for Business and Economics 8th edn, Prentice Hall, NJ.
Neter J, Wasserman W (1974) Applied linear statistical models. Richard D Irwin, Inc., Homewood, III.
Pappis CP, Karacapilidis NI (1993) A comparative assessment of measures of similarity of fuzzy values. Fuzzy Sets and Systems 56(2):171-174
Parsons S (1996) Current approaches to handling imperfect information in data and knowledge bases. IEEE Transactions on Knowledge and Data Engineering 8(3):353-372
Peña-Reyes CA, Sipper M (2000) A fuzzy genetic approach to breast cancer diagnosis, Artif. Intell. Med. 17:131-155
Ross TJ (1995) Fuzzy logic with engineering applications. McGraw-Hill Inc Waugh F (1928) Quality Factors Influencing Vegetable Prices. Journal of Farm Economics 10:185-196.
Setiono R (2000) Generating concise and accurate classification rules for breast cancer diagnosis, Artif. Intell. Med. 18:205-219
Subtil P, Mouaddib N and Foucaut O (1996) A fuzzy information retrieval and management system and its applications, Proceedings of the ACM symposium on Applied Computting, 1906:537-541
The New York Times Company (2006) http://databases.about.com/cs/sql/a/aa042803a.htm
Wang JW, Cheng CH, Forecasting null value using Fuzzy partition method for relational database (submitted to Intelligent Data Analysis).
Wang JW, Cheng CH (2006a) An Efficient Method for Estimating Null Values in Relational Databases. Knowledge and Information Systems. Accepted.
Wang JW, Cheng CH (2006b) Fuzzy Clustering-based on aggregate attribute method,” IEA/AIE, France, Lecture Notes in Computer Science, 4031: 478-487.
Wang JW, Cheng CH and Chang WT (2005) Partitional approach for estimating null value in relational database, AI 2005, Australia. Lecture Notes in Computer Science, 3809:1213-1216.
Wang JW, Chang JR and Cheng CH (2006) Flexible Fuzzy OWA Querying Method for Hemodialysis Database, Soft Computing. 10(11):1031-1042.
Waugh F (1928) Quality Factors Influencing Vegetable Prices, Journal of Farm Economics. 10:185-196
Young HP (1988) Condorcet’s theory of voting. American Political Science Review. 82:1231-1244
Zadeh LA (1965) Fuzzy sets. Information and Control 8:338-353
Zadeh LA (1973) The concept of a linguistic variable and its application to approximate reasoning. Memorandum ERL-M 411, Berkeley, October 1973
Zaniolo C (1984) Database Relations with Null Values. Reprinted from journal of computer and system sciences 28(1):142-166
Zimmermann HJ (1991) Fuzzy Set Theory and it Applications, 4th ed. Kluwer Academic Publishers