SINIFLANDIRMA KURALLARININ ÇIKARIMI İÇİN ETKİN VE HASSAS YENİ BİR YAKLAŞIM

Yıl 2014, , 0 - , 30.09.2014

Murat Köklü , Humar Kahramanlı , Novruz Allahverdi

https://doi.org/10.17341/gummfd.89433

Cited By: 3

Öz

Bu çalışmada çok sınıflı verilerden kural çıkarımı için yeni bir yöntem önerilmiştir. Önerilen metot 4 farklı veri kümesi üzerinde uygulanmıştır. Ayrık ve gerçel nitelikler farklı şekilde kodlanmıştır. Ayrık nitelikler ikili olarak, gerçek nitelikler ise, iki gerçel değer kullanılarak kodlanmıştır. Bu değerler kuralları oluşturan niteliklerin değer aralıklarının orta noktası ve genişlemesidir. Kural çıkarım işlemi için sınıflandırma başarısı uygunluk fonksiyonu olarak kullanılmıştır. Uygunluk fonksiyonunun optimizasyonu amacıyla Yapay Bağışıklık Sistemi (YBS) algoritması olan CLONALG kullanılmıştır. Önerilen yöntemi uygulamak için Süsen çiçeği (Iris), Şarap (Wine), Cam Kimliklendirme (GlassIdentification) ve Deniz Kabuğu (Abalone) veri kümeleri kullanılmıştır. Veriler Irvine California Üniversitesi (UCI) makine öğrenmesi veri tabanından elde edilmiştir. Önerilen metot kullanılarak Süsen çiçeği için %100, Şarap için %99,44, Cam kimliklendirme için %77,10 ve Deniz Kabuğu için %62,59 doğruluk elde edilmiştir. Diğer yöntemlerle kıyaslandığında önerilen yöntem kullanılarak daha başarılı sonuçlar elde edildiği ve hesaplama karmaşıklığı açısından da avantajlı olduğu görülmüştür.

Anahtar Kelimeler

Kural çıkarımı, Gerçel değer kodlaması, Çok sınıflı problemler.

Kaynakça

• Kahramanlı H., and Allahverdi N., “Rule Extraction from Trained Adaptime Neural Networks Using ArtificialİmmuneSystems”, Expert Systems With Applications, Sayı 36, pp, 1513-1522, 2009.
• Gallant, S. I., “Connection Expert Systems”,Communications of the ACM, 31(2), pp.152–169, 1998.
• Towell, G. G., and Shavlik, J., “Extracting Refined Rules From Knowledge-Based Neural Networks”. Machine Learning, 13, pp.71–101, 1993.
• Lu, H., Setiono, R., and Liu, H., “Effective Data Mining Using Neural Networks”,IEEE Transactions on Knowledge and Data Engineering, 8(6), pp.957–961, 1996.
• Keedwell, E., Narayanan, A., and Savic, D., “Evolving rules from neural networks trained on continuous data”,Evolutionary Computation Congress,Proceedings Of The Congress On Evolutionary Computation, 1, pp.639-645, 2000a.
• Keedwell, E., Narayanan, A., and Savic, D., “Creating Rules From Trained Neural Networks Using Genetic Algorithms”International Journal of Computers Systeming Signals (IJCSS), 1(1), pp.30–42, 2000b.
• Aliev R.A., Aliev R.R., Guirimov B. and Uyar K., “Dynamic Data Mining Technique for Rules Extraction in a Process of Battery Charging”, Applied Soft Computing, 8, pp. 1252–1258, 2008.
• Ang J.H., Tan K.C. and Mamun A.A., “An Evolutionary Memetic Algorithm For Rule Extraction”, Expert Systems with Applications, Volume 37, 2, pp.1302-1315, 2010.
• Papageorgiou E.I., “A New Methodology for Decisions in Medical Informatics Using Fuzzy Cognitive Maps Based on Fuzzy Rule-Extraction Techniques”, Applied Soft Computing, 11, pp.500–513, 2011.
• Rodríguez M., Escalante D. M., and Peregrín A., “Efficient Distributed Genetic Algorithm for Rule Extraction”, Applied Soft Computing, Vol.11, 1, pp. 733-743,2011.
• Sarkar, B.K., Sana S.S. and Chaudhuri K., “A Genetic Algorithm-Based Rule Extraction System”,Applied Soft Computing 12,pp.238–254, 2012.
• Özbakır L., Baykasoğlu A. and Kulluk S., “A Soft Computing-Based Approach for Integrated Training and Rule Extraction From Artificial Neural Networks: DIFACONN-Miner”, Applied Soft Computing 10,pp.304–317, 2010.
• Baykasoglu A., Saltabaş A., Taşan A.S. and Subulan K., “Yapay Bağışıklık Sisteminin Çoklu Etmen Benzetim Ortamında Realize Edilmesi ve Gezgin Satıcı Problemine Uygulanması” Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, Cilt 27, No 4, 901-909, 2012.
• Seredynski, F., and Bouvry, P., “Anomaly Detection in TCP/IP Networks Using Immune Systems Paradigm”,Computer Communications, 30, pp.740–749, 2007.
• Kalinli, A., and Karaboga, N., “Artificial Immune Algorithm for IIR Filter Design”,Engineering Applications of Artificial Intelligence, 18, pp.919–929, 2005.
• Musilek, P., Lau, A., Reformat, M., & Wyard-Scott, L., “Immune Programming”,Information Sciences, 176, pp.972–1002, 2006.
• Kumar, A., Prakash, A., Shankar, R., & Tiwari, M. K., “Psychoclonal Algorithm Based Approach to Solve Continuous Flow Shop Scheduling Problem”,Expert System with Applications, 31, pp.504–514, 2006.
• De Castro, L. N., & Timmis, J., “Artificial Immune Systems: ANew Computational Intelligence Approach”. UK: Springer, 2002.
• Hart, E. and Timmis, J., “Application Areas of AIS: The past, The Present and The Future”, International Conference on Artificial Immune Systems (ICARIS), Canada, LNCS 3627, pp. 483-497, 2005.
• Timmis, J., Hone, A., Stibor, T. and Clarck, E., “Theoretical advances in artficial immune systems”, Theoretical Computer Science, 403, pp. 11-32, 2008.
• Brownlee, J., “Clonal Selection Theory & CLONALG the Clonal Selection Classification Algorithm (CSCA)”, Tecnical Report, 2005.
• Parpinelli, R.S., Lopes, H.S., Freitas, A.A., “An Ant Colony Based System For Data Mining: Applications To Medical Data”,Proceedings of the Genetic and Evolutionary Computation Conference, San Francisco, California, pp.791–798, 2001.
• Frank, A. and Asuncion, A., “UCI Machine LearningRepository”,[http://archive.ics.uci.edu/ml], Irvine, CA: University of California, School of Information and Computer Science, Last accessed:February, 2011.
• Clark D., Schreter Z. and Adams A., "A Quantitative Comparison of Dystal and Backpropagation", Submitted to the Australian Conference on Neural Networks,1996.
• Rijnbeek P.R. and Kors J.A., “Finding a Short and Accurate Decision Rule in Disjunctive Normal Form by Exhaustive Search”, Machine Learning, 80: 33–62, DOI 10.1007/s10994-010-5168-9, 2010.
• Nojima Y., Ishibuchi H. And Kuwajima I., “Parallel Distributed Genetic Fuzzy Rule Selection (FDGFRS)”, Soft Computing, 13, pp.511–519, 2009.
• Ful X. and Wang L., “A Rule Extraction System with Class-Dependent Features”, Studies in Fuzziness and Soft Computing, Volume 163/2005, pp.79-99, DOI: 10.1007/3-540-32358-9_5, 2005.
• Özbakır, L., Baykasoğlu, A., Kulluk, S., Yapıcı, H., “TACO-miner: An Ant Colony Based Algorithm for Rule Extraction from Trained Neural Networks”, Expert Systems with Applications, 36, 12295-12305, 2009.
• Özbakır, L.,Delice,Y., “Exploring comprehensible classification rules from trained neural Networks integrated with a time-varying binary particle swarm optimizer”, Engineering Applications of Artificial Intelligence, 24, 491-500, 2011.
• Browne C., Duntsch I. and Gediga G., “IRIS Revisited: A Comparison Of Discriminant and Enhanced Rough Set Data Analysis”. In: L. Polkowski and A. Skowron, eds. Rough Sets in Knowledge Discovery, vol. 2. Physica Verlag, Heidelberg, 345-368, 1998.
• Weiss S.M. and Kapouleas, I., "An Empirical Comparison of Pattern Recognition, Neural Nets and Machine Learning Classification Methods", in: J.W. Shavlik and T.G. Dietterich, Readings in Machine Learning, Morgan Kauffman Publ, CA 1990.
• Duch W, Adamczak R, Grąbczewski K., “A New Methodology of Extraction, Optimization and Application of Crisp and Fuzzy Logical Rules”. IEEE Transactions on Neural Networks, 12, pp.277-306, 2001.
• Nauck D., Nauck U. and Kruse R., “Generating Classification Rules with the Neuro-Fuzzy System NEFCLASS”,Proc. Biennial Conf. of the North American Fuzzy Information Processing Society (NAFIPS'96), Berkeley, 1996.
• Pal N. R. and Chakraborty S., “Fuzzy Rule Extraction From ID3-Type Decision Trees for Real Data”, IEEE Trans. Syst., Man, Cybern. B. 31 pp. 745-754, 2001.
• Wang S., Wu G. and Pan J., “A Hybrid Rule Extraction Method Using Rough Sets and Neural Networks”, Lecture Notes in Computer Science, 2, 4492, pp. 352–361, 2007.
• Castellano G., Fanelli A. M., and Mencar C., “An empirical risk functional to improve learning in a neuro-fuzzy Classifier”,IEEE Trans. Syst., B. 34 pp. 725-731, 2004.
• Kim D.W., Park J.B., and Joo Y.H., “Design of Fuzzy Rule-Based Classifier: Pruning and Learning”, Fuzzy Systems and Knowledge Discovery Lecture Notes in Computer Science, 3613, pp. 416–425, Springer, 2005.
• Pappa G.L. and Freitas A. A., “Evolving Rule Induction Algorithms with Multi-Objective Grammar-Based Genetic Programming”, Knowledge and Information Systems, 19, pp.283–309. DOI 10.1007/s10115-008-0171-1, 2009.
• De Castro, L. N.,, "Learning and Optimization Using the Clonal Selection Principle",IEEE Transactions On Evolutionary Computation, Vol. 6, no. 3, June, 2002.
• C 4.5 Lecture Notes, [http://www.sts.tu-harburg.de/teaching/ss-09/ml-sose-09/03-Decision-Tree-c45.pdf], Hamburg University of Technology, Lecture Notes in Electronic Science,Last accessed,: Semptember, 2013.
• Huysmans J., Baesens B. and Vanthienen J, “Using Rule Extraction to Improve the Comprehensibility of Predictive Models”, Open Access publications from Katholieke Universiteit Leuven with number urn:hdl:123456789/121060, KBI-0612, 2006.7