Endüktif öğrenme algoritmalarının kural üretme yöntemleri ve performanslarının karşılaştırılması

Bilgi Çağı" ve "Bilgi Toplumu" gibi terimlerin sıklıkla kullanıldığı günümüzde, bilginin önemi daha açık bir şekilde ortaya çıkmaktadır. Bilginin önemi arttığı oranda o bilgiye ulaşabilmeyi sağlayan sistemlerin de önemi artmaktadır. Bilgisayar teknolojisindeki büyük gelişmeler sayesinde, dünyanın herhangi bir yerinde üretilen bilginin sayısal hale getirilerek saklanması ve o bilgiye dünyanın herhangi bir yerinden çok kısa sürede erişim mümkün olmaktadır. Bununla birlikte programlama dillerindeki büyük gelişmeler sayesinde bilgiyi işlemek ve istenen bilgiye erişmek de kolaylaşmaktadır. Bu çalışmada, bilgiyi elde etmek amacıyla kullanılan Endüktif Öğrenme teknikleri anlatılacak ve bu alanda geliştirilen algoritmalar karşılaştırılacaktır.

Rule generation methods of inductive learning algorithms and comparison of their performances

Frequent emphasis on the phrases such as "Information Age" and "Information Society" clearly expresses the importance of knowledge in our daily life. As the importance of knowledge increases, so does the need for tools to reach and retrieve the knowledge. Thanks to great developments in the computer technology, it is possible to store the information generated miles away in an electronic format and retrieve it back quickly at any location in the world, when needed. In addition, rapid developments in programming languages also made it easy to process the information and access it in case of need. In this study, inductive learning techniques which are used to acquire information will be explained and the algorithms developed for such purposes, will be compared with each other.

PDF

___

[1]. FEIGENBAUM, E. A., "Expert System in the 1980s", in Infotech State of the Art Report on Machine Intelligence, Ed:A. Bond, Maidenhead, Pergamon, Infotch, 1981.
[2]. BOSE, I., MAHAPATRA K.R., "Business Data Mining - A Machine learning Perspective", Information & Management 39, 211-225, 2001.
[3]. VAPNIK V.N., CHERVONENKIS A.Y., "On the uniform convergence of relative frequencies of events to their probabilities", Theory of Probability and Its Applications, 16(2), 264-280, 1971.
[4]. VALIANT, L.G., "A theory of the learnable", Communications of the ACM, Vol.27, 1134-1142; or, Readings in Machine Learning, (1990) Shavlik, J.W. and Dietterich, T.G. (Eds), Morgan Kaufmann, San Mateo, CA, 192-200, 1984.
[5]. OBLOW, E.M., "Implementing Valiant's learnability theory using random sets", Machine Learning, 8(1), Kluwer Academic Publishers, Boston, 45-73, 1992.
[6]. KRZYSZTOF J. C., "An Algorithm Which Learns Multiple Covers Via Integer Linear Programming Part I: the CL1LP2 algorithm", Kybernetes, Vol. 24 "No. 2, pp. 29-50, MCB University Press, 0368-492X, 1995.
[7]. MICHALSKI, R.S., "A theory and methodology of inductive learning", Machine Learning - An Artificial Intelligence Approach, Michalski; R.S., Carbonell, J.G. and Mitchell, T.M. (Eds), Morgan Kaufmann, Los Altos, CA, 83-134, 1983.
[8]. MITCHELL, T.M., "The need for biases in learning generalizations", Readings in Machine Learning, Shavlik, J.W. and Dietterich, T.G. (Eds), Morgan Kaufmann, San Mateo, CA, 184-191, 1990.
[9]. QU1NLAN, J.R., "Learning efficient classification procedures and their application to chess end games", Machine Learning - An Artificial Intelligence Approach, Eds: Michalski; R.S., Carbonell, J.G. and Mitchell, T.M. (Eds), Tioga Publishing Co, Palo Alto. CA, 463-482, 1983.
[10]. DIETTERICH, T.G., "Limitations on inductive learning", Proceedings of the 6th International Workshop on Machine Learning (89 ML), Ithaca, NY. and Segre, A.M. (Ed), Morgan Kaufmann, San Mateo, CA, 124-128, 1989.
[11]. BOLC, L., CYTOWSKI, J., "Search Methods for Artificial Intelligence", Academic Press, London, 1992.
[12]. HAUSSLER, D. "Quantifying inductive bias: AI learning algorithms and Valiant's learning framework", Artificial Intelligence, 36, 177-221; reproduced in:, Readings in Machine learning, 1990.
[13]. WANG, X., "Inductive Learning Algorithms", Ph.D. Thesis, University of Wales Cardiff, 1997.
[14]. MICHALSKI, R.S., KODRATOFF, Y., "Research in machine learning: recent progress, classification of methods, and future directions", Machine Learning Vol.3, Morgan Kaufmann, San Mateo, CA, 3-30, 1990.
[15]. CENDROWSKA, J., "Knowledge Acquisition for Expert Systems: Inducing Modular Rules from Examples", PhD Thesis, The Open University, 1990.
[16]. BRAMER, M.A., "Automatic Induction of Classification Rules from Examples Using N-Prism",. Research and Development in Intelligent Systems XVI. Springer-Verlag, pp. 99-121, 2000.
[17]. HUNT, E. B., MARIN, J., STONE, P. J., "Experiments in induction", Academic Pess, New York, 1966.
[18]. QUINLAN, J.R., "Induction of decision trees", Machine Learning Vol.1, Kluwer Academic Publishers, Boston, 81-106; reproduced in:. Readings in Machine learning, 1990.
[19]CHENG, J., et al., "Improved decision trees: A generalized version of 1D3", Proceedings of the Fifth international conference on Machine Learning, Ann Arbor, Michigan, 100-106, 1988.
[20]SCHLIMMER, J.C., FISHER, D.H., "A case study of incremental concept induction", AAAI 86- Proceedings of the 5th National Conference on Artificial Intelligence, Philadelphia, PA, 496-501, 1986.
[21]UTGOFF, P.E., "ID5:An incremental ID3", Proceedings of the Fifth International Conference on Machine Learning, The University of Michigan, 107-120, 1988.
[22]UTGOFF, P.E., "Incremental induction of decision trees", Machine Learning Vol.4, 161-186, 1989.
[23]QUINLAN, J.R., "C4.5: Programs for Machine Learning", Morgan Kaufmann, San Mateo, CA, 1993.
[24]BREIMAN, L.,et al. "Classification and Regression Trees", Wadsworth International Group, Belmont, California, 1984.
[25]CRAWFORD,S.L."Extensions to the CART algorithm", Machine Learning and Uncertain Reasoning - Knowledge-Based Systems, Vol. 3, Gaines, B and Boose, J. (Eds), Academic Press, London, 15-35, 1990.
[26]ZHONG, N., DONG, J., OHSUGA, S., "Rule discovery by soft induction techniques", Neurocomputing 36, p: 171-204,2001.
[27]MICHALSKI, R.S., "Synthesis of optimal and quasi-optimal variable-valued logic formulas", Proceeding of the 1975 Int. Symposium on Multiple-Valued Logic, Bloomington, Indiana, 76-87, 1975.
[28]AKSOY, M.S., "New Algorithms for Machine Learning", Thesis of PhD, University of Walles, Cardiff, United Kingdom, 1993. [29]. TOLUN, M. R., ABU-SOUD S.M., "ILA.An Inductive Learning Algorithm For Rule Extraction", Expert Systems With Applications, Vol: 14, p:361-370, 1998.
[30]. TOLUN, M. R., et al., "Improved Rule Discovery Performance on Uncertainty", The Second Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD-98), Melbourne, Australia, 15-17 April 1998.
[31]. AKGÖBEK, Ö., "Endüktif Öğrenmede Bilgi Kazanımı için Yeni Algoritmalar", Doktora Tezi, Sakarya Üniversitesi, Adapazarı, 2003.
[32]. BLAKE, C.L., MERZ, C.J., "UCI Repository of Machine Learning Databases", [http://ftp.ics.uci.edu /pub/ml-repos/machine-learning-databases/], 1998.
[33]. SGI Standard Template Library Programmer's Guide, Silicon Graphics Inc., http://www.sgi. com/tech/mcl/db., 1996.
[34], BRAMER, M.A., "Using J-pruned to reduce overfitting in classification trees", Knowledge-Based Systems, Vol: 15,301 -308, 2002.
[35]. WU, X., "Rule Induction with Extension Matrices", Journal of the American Society for Information Science, Volume 49, Vol 5, 435-454, 1998.
[36]. HAMILTON, H. J., et al., "RIAC:A Rule Induction Algorithm Based on Approximate Classification Technical Report", S4S 0A2, CS-96-06, ISSN 0828-3494 ISBN 0-7731-0321-X, 35-37, 1996.
[37] THRUN, S.B., BALA, J.,"The MONK's Problems A Performance Comparison of Different Learning Algorithms", Carnegie Mellon University, CMU-CS-91-197, 1991.
[38]. AN, A., "Learning Classification Rules From Data", Computers & Mathematics with Applications, Vol 45, Issues 4-5, 737-748, 2003.
[39]. FOURNIER, D., CREMILLEUX, B., "A Quality Index For decision Tree Pruning", Knowledge-Based System 15,37-43,2002.