Genelleştirilmiş ilginç sınıflandırma kurallarının yapay kimyasal reaksiyon optimizasyon algoritması ile keşfi

Sınıflandırma kural madenciliği en çok çalışılan veri madenciliği problemlerindendir. Bu makalede birdenfazla sınıflama etiketi bulunan ve tahmin edilecek birden fazla hedef niteliğin olduğu veritabanlarında,sınıflandırma kural madenciliğinin daha genel bir hali olan kompleks ve fazla çalışılmamış genelleştirilmişkural keşfi problemi için ilk kez kimya tabanlı Yapay Kimyasal Reaksiyon Optimizasyon Algoritması(YAKROA) kullanılmıştır. İlginçlik kriteri de eklenerek; algoritmanın keşfedeceği kuralların sadece doğru ve anlaşılabilir değil aynı zamanda ilginç, beklenmedik ve sürpriz olması için de gerekli düzenlemeleryapılmıştır. Farklı amaçlar doğrultusunda değişik veritabanlarında farklı kurallar kümesi, esnek bir şekildealgoritmadaki temsil biçimi ve amaç fonksiyonunun düzenlenmesiyle bulunmuştur. Farklı özellikte halkaaçık gerçek veritabanlarında YAKROA'nın sınıflandırma kural keşfi problemindeki performansı genetikalgoritma, parçacık sürü optimizasyon algoritması ve karınca koloni optimizasyon algoritması ile karşılaştırılmıştır. Deneysel sonuçlardan, YAKROA'nın veri madenciliğinin bu özel alanındakiperformansının umut verici olduğu görülmüştür. YAKROA'nın farklı veri madenciliği problemleri;özellikle birliktelik kurallarının keşfi, kümeleme kurallarının keşfi, ardışık örüntü keşfi vb. için etkili birçözüm yöntemi olarak kullanılabileceği öngörülmektedir.

Mining of generalized interesting classification rules with artificial chemical reaction optimization algorithm

Classification rules mining is one of the most studied data mining problems. In this article, chemistry basedArtificial Chemical Reaction Optimization Algorithm (ACROA) has been for the first time used for miningof generalized classification rules, which is a complex and no well researched generalized variant ofclassification rules mining where there is more than one goal attribute to be predicted. Furthermore,interestingness measure has been added by performing the adaptations to the algorithm in order to make the rules mined by the algorithm not only accurate and comprehensible but also interesting, surprising, andunexpected. Different rule sets within different databases satisfying different objectives have been flexiblymined by adapting the representation scheme and objective function. Performance of ACROA inclassification rules mining within different real public databases have been compared that of geneticalgorithm, particle swarm optimization algorithm, and ant colony optimization algorithm. It has shown that performance of ACROA within this special task of data mining is promising. ACROA can be an efficientsolution method for different data mining tasks such as association rules mining, clustering rules mining,sequential pattern mining, and etc.

PDF

___

1. Han J., Kamber M., Pei J., Data Mining: Concepts and Techniques, The Morgan Kaufmann Series in Data Management Systems, 3, 744, 2011
2. Setiono R., Azcarraga A., Hayashi Y., Using Sample Selection to Improve Accuracy and Simplicity of Rules Extracted from Neural Networks for Credit Scoring Applications, International Journal of Computational Intelligence and Applications, 14 (4), 1550021, 2015.
3. Quinlan J.R., C4.5: Programs for Machine Learning, Morgan Kaufmann, 1992.
4. Breiman L., Friedman J.H., Olshen R.A., Stone C. J., Classification and Regression Trees, Wadsworth, 1984.
5. Berzal F., Cubero J.C., Sánchez D., Serrano J.M., Art: A Hybrid Classification Model, Machine Learning, 54 (1), 67-92, 2004.
6. Dai Q., Zhang C. Wu H., Research of Decision Tree Classification Algorithm in Data Mining, International Journal of Database Theory and Application, 9 (5), 1-8, 2016.
7. Alatas B., Akin E., Sınıflandırma Kurallarının Karınca Koloni Algoritmasıyla Keşfi, ASYU Akıllı Sistemlerde Yenilikler ve Uygulamaları Sempozyumu, Istanbul, 62- 66, 2006.
8. Aggarwal C.C., Data Classification: Algorithms and Applications, CRC Press, 2014.
9. Sharma P., Discovery of Classification Rules Using Distributed Genetic Algorithm, Procedia Computer Science, 46, 276-284, 2015.
10. Panda M., Abraham A., Hybrid Evolutionary Algorithms for Classification Data Mining, Neural Computing and Applications, 26 (3), 507-523, 2015.
11. Gundogan K.K., Alatas B., Karci A., Mining Classification Rules by Using Genetic Algorithms with Non-Random Initial Population and Uniform Operator, Turkish Journal of Electrical Engineering & Computer Sciences, 12 (1), 43-52, 2004.
12. Yang L., Li K., Zhang W., Ke Z., Ant Colony Classification Mining Algorithm Based on Pheromone Attraction and Exclusion, Soft Computing, 1-13, 2016.
13. Liang Z., Sun J., Lin Q., Du Z., Chen J., Ming Z., A Novel Multiple Rule Sets Data Classification Algorithm Based on Ant Colony Algorithm, Applied Soft Computing, 38, 1000-1011, 2016.
14. Asadi S., Shahrabi J., ACORI: A Novel ACO Algorithm for Rule Induction, Knowledge-Based Systems, 97, 175-187, 2016.
15. Alatas B., Akin E., Multi-Objective Rule Mining Using a Chaotic Particle Swarm Optimization Algorithm, Knowledge-Based Systems, 22 (6), 455-460, 2009.
16. Tapkan P., Özbakır L., Kulluk S., Baykasoğlu, A., A Cost-Sensitive Classification Algorithm: BEE-Miner, Knowledge-Based Systems, 95, 99-113, 2016.
17. Celik M., Karaboga D., Koylu F., Artificial Bee Colony Data Miner (ABC-Miner), IEEE International Symposium on Innovations in Intelligent Systems and Applications (INISTA), 96-100, 2011.
18. Al-Sheshtawi K.A., Abdul-Kader H.M., Elsisi A.B., A Novel Artificial Immune Clonal Selection Classification and Rule Mining with Swarm Learning Model, Connection Science, 25 (2-3), 75-127, 2013.
19. Köklü M., Kahramanlı H., Allahverdi, N., A New Accurate and Efficient Approach to Extract Classification Rules, Journal of the Faculty of Engineering and Architecture of Gazi University, 29 (3), 477-486, 2014.
20. Akyol S., Alatas, B, Automatic mining of accurate and comprehensible numerical classification rules with cat swarm optimization algorithm, Journal of the Faculty of Engineering and Architecture of Gazi University, 31 (4), 839-857, 2016.
21. Alatas B., Akin E., Mining Fuzzy Classification Rules Using an Artificial Immune System with Boosting, Advances in Databases and Information Systems, 283- 293, Springer Berlin Heidelberg, 2005.
22. Alatas B., Akin E., FCACO: Fuzzy Classification Rules Mining Algorithm with Ant Colony Optimization, Advances in Natural Computation, 787- 797, Springer Berlin Heidelberg, 2005.
23. Pourpanah F., Lim C.P., Saleh J.M., A Hybrid Model of Fuzzy ARTMAP and Genetic Algorithm for Data Classification and Rule Extraction, Expert Systems with Applications, 49, 74-85, 2016.
24. Kar A.K., Bio Inspired Computing-A Review of Algorithms and Scope of Applications, Expert Systems with Applications, 59, 20-32, 2016.
25. Akyol S., Alatas B., Plant Intelligence Based Metaheuristic Optimization Algorithms, Artificial Intelligence Review, 1-46, 2016.
26. Blum C., Raidl G. R., Further Hybrids and Conclusions, Hybrid Metaheuristics, 127-136, Springer International Publishing, 2016.
27. Ozbay F.A., Alatas B., Review of Musics based Computational Intelligence Algorithms, 1st International Conference on Engeneering Technology and Applied Sciences, Afyon Kocatepe University, 663-669, 2016.
28. Bingol H., Alatas B., Chaotic League Championship Algorithms, Arabian Journal for Science and Engineering, 41 (12), 1-25, 2016.
29. Alatas B, ACROA: Artificial Chemical Reaction Optimization Algorithm for Global Optimization, Expert Systems with Applications, 38 (10), 13170- 13180, 2011.
30. Alatas B., A Novel Chemistry Based Metaheuristic Optimization Method for Mining of Classification Rules, Expert Systems with Applications, 39 (12), 11080-11088, 2012.
31. Alatas B., Karci A., Genetik Algoritmalarda Düzenli Popülasyon ve Düzenli Operatör, Afyon Kocatepe Üniversitesi Fen Bilimleri Dergisi, 3 (1-2), 11-26, 2003.
32. Demir M., Karci A., Veri Kümelemede Fidan Gelişim Algoritmasının Kullanılması, 12. Elektrik, Elektronik, Bilgisayar, Biyomedikal Mühendisliği Ulusal Kongresi ve Fuarı, Eskisehir, 14-18 2007.
33. Karci A., Alatas A., Akin E., Fidan Gelişim Algoritması, ASYU Akıllı Sistemlerde Yenilikler ve Uygulamaları Sempozyumu, Istanbul, 57-61, 2006.
34. Karci A., Theory of Saplings Growing up Algorithm, Lecture Notes in Computer Science, 4431, 450-460, 2007.
35. Karci A., Alatas B., Thinking Capability of Saplings Growing Up Algorithm, Lecture Notes in Computer Science, 4224, 386-393, 2006.
36. Eberbach E., The Role of Completeness in Convergence of Evolutionary Algorithms, IEEE Congress on Evolutionary Computation, 2, 1706-1713, 2005.
37. Lichman M., UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science, Erişim tarihi 1 Temmuz 2014.
38. Freitas A.A., On Objective Measures of Rule Surprisingness, 2nd European Symposium on Principles of Data Mining and Knowledge Discovery (PKDD-98), Lecture Notes in Artificial Intelligence, 1510, 1-9, 1998