RULES-IT: incremental transfer learning with RULES family

Abstract

In today’s world of excessive development in technologies, sustainability and adaptability of computer applications is a challenge, and future prediction became significant. Therefore, strong artificial intelligence (AI) became important and, thus, statistical machine learning (ML) methods were applied to serve it. These methods are very difficult to understand, and they predict the future without showing how. However, understanding of how machines make their decision is also important, especially in information system domain. Consequently, incremental covering algorithms (CA) can be used to produce simple rules to make difficult decisions. Nevertheless, even though using simple CA as the base of strong AI agent would be a novel idea but doing so with the methods available in CA is not possible. It was found that having to accurately update the discovered rules based on new information in CA is a challenge and needs extra attention. In specific, incomplete data with missing classes is inappropriately considered, whereby the speed and data size was also a concern, and future none existing classes were neglected. Consequently, this paper will introduce a novel algorithm called RULES-IT, in order to solve the problems of incremental CA and introduce it into strong AI. This algorithm is the first incremental algorithm in its family, and CA as a whole, that transfer rules of different domains to improve the performance, generalize the induction, take advantage of past experience in different domain, and make the learner more intelligent. It is also the first to introduce intelligent aspects into incremental CA, including consciousness, subjective emotions, awareness, and adjustment. Furthermore, all decisions made can be understood due to the simple representation of repository as rules. Finally, RULES-IT performance will be benchmarked with six different methods and compared with its predecessors to see the effect of transferring rules in the learning process, and to prove how RULES-IT actually solved the shortcoming of current incremental CA in addition to its improvement in the total performance.

Keywords

incremental learning / transfer learning / covering algorithms / RULES family / inductive learning

References

[1] Yudkowsky E. Levels of organization in general intelligence. In: Goertzel B, Pennachin C, eds. Artificial General Intelligence. Berlin: Springer, 2007, 389−501

[2] Pennachin C, Goertzel B. Contemporary approaches to artificial general intelligence. In: Goertzel B, Pennachin C, eds. Artificial General Intelligence. SpringerLink, 2007, 1−28

[3] Shita M, Gilman N, Deighton N, Pedersen M, Dodsworth C, Oana J. Kimera Systems, 2013.

[4] Aksoy M S, Mathkour H, Alasoos B A. Performance evaluation of RULES-3 induction system for data mining. International Journal of Innovative Computing, Information and Control, 2010, 6: 3339−3346

[5] Kotsiantis S B. Supervised machine learning: a review of classification techniques. Informatica (03505596), 2007, 31: 249−268

[6] Cios K J, Swiniarski R W, Pedrycz W, Kurgan L A, Cios K, Swiniarski R, Kurgan L. Supervised learning: decision trees, rule algorithms, and their hybrids. Data Mining, eds. US: Springer, 2007, 81−417

[7] Birzniece I. The use of inductive learning in information systems. In: Proceedings of the 16th International Conference on Information and Software Technologies. 2010, 95−101

[8] Qin Z, Wan T. Hybrid bayesian estimation tree learning with discrete and fuzzy labels. Frontiers of Computer Science, 2013, 1−12

[9] Kurgan L A, Cios K J, Dick S. Highly scalable and robust rule learner: performance evaluation and comparison. IEEE Systems, Man, and Cybernetics—Part B: Cybernetics, 2006, 36: 32−53

[10] Pan S J, Yang Q. A Survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22: 1345−1359

[11] Alcalá-Fdez J, Sánchez L, García S, Jesus M J d, Ventura S, Garrell J M, Otero J, Romero C, Bacardit J, Rivas V M, Fernández J C, Herrera F. KEEL: a software tool to assess evolutionary algorithms to data mining problems. Soft Computing, 2009, 13: 307−318

[12] Alcalá-Fdez J, Fernandez A, Luengo J, Derrac J, García S, Sánchez L, Herrera F. KEEL data-mining software tool: data set repository, integration of algorithms and experimental analysis framework. Journal of Multiple-Valued Logic and Soft Computing, 2011, 17: 255−287

[13] Efron B, Tibshirani R. An Introduction to the Bootstrap. USA: Chapman& Hall, 1993

[14] Aksoy M S. A review of rules family of algorithms. Mathematical and Computational Applications, 2008, 13: 51−60

[15] Pham D T, Aksoy M S. RULES: a simple rule extraction system. Expert Systems with Applications, 1995, 8: 59−65

[16] Pham D T, Aksoy M S. An algorithm for automatic rule induction. Artificial Intelligence in Engineering, 1993, 8: 277−282

[17] Pham D T, Aksoy M S. A new algorithm for inductive learning. Journal of Systems Engenering, 1995, 5: 115−122

[18] Pham D T, Dimov S S. The RULES-3 plus inductive learning algorithm. In: Proceedings of the 3rd World Congress on Expert Systems. 1996, 917−924

[19] Mathkour H I. RULES3-EXT improvement on RULES-3 induction algorithm. Mathematical and Computational Applications, 2010, 15(3): 318−324

[20] Pham D T, Dimov S S. An algorithm for incremental inductive learning. Journal of Engineering Manufacture, 1997, 211: 239−249

[21] Pham D T, Soroka A J. An immune-network inspired rule generation algorithm (RULES-IS). In: Proceedings of the 3rd Virtual International Conference on Innovative Production Machines and Systems. 2007

[22] Pham D T, Bigot S, Dimov S S. RULES-5: a rule induction algorithm for classification problems involving continuous attributes. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2003, 217(12): 1273−1286

[23] Pham D T, Bigot S, Dimov S S. RULES-F: a fuzzy inductive learning algorithm. Proceedings of the Institution ofMechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2006, 220: 1433−1447

[24] Bigot S. A new rule space representation scheme for rule induction in classification and control applications. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 2011, 225(7): 1018−1038

[25] Pham D T, Afify A A. RULES-6: a simple rule induction algorithm for supporting decision making. In: Proceedings of the 31st Annual Conference of IEEE Industrial Electronics Society. 2005, 6

[26] Shehzad K. EDISC: a class-tailored discretization technique for rulebased classification. IEEE Transactions on Knowledge and Data Engineering, 2012, 24: 1435−1447

[27] Afify A A, Pham D T. SRI: a scalable rule induction algorithm. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2006, 220: 537−552

[28] ElGibreen H, Aksoy M S. RULES-TL: a simple and improved RULES algorithm for incomplete and large data. Journal of Theoretical and Applied Information Technology, 2013, 47

[29] ElGibreen H, Aksoy M S. Multi model transfer learning with RULES family. In: Proceedings of the 2013 International Conference on Machine Learning and Data Mining. 2013, 42−56

[30] Ramon J, Driessens K, Croonenborghs T. Transfer learning in reinforcement learning problems through partial policy recycling: machine learning. In: Kok J, Koronacki J, Mantaras R, Matwin S, Mladenic D, Skowron A, eds. European Conference on Machine Learning. Berlin: Springer, 2007, 4701: 699−707

[31] Taylor M, Suay H B, Chernova S. Integrating reinforcement learning with human demonstrations of varying ability. In: Proceedings of the 10th International Conferance of Autonomous Agents and Multiagent Systems. 2011, 617−624

[32] Mahmud M. On universal transfer learning algorithmic learning theory. In: Hutter M, Servedio R, Takimoto E, eds. Berlin: Springer, 2007, 4754: 135−149

[33] Taylor M, Kuhlmann G, Stone P. Accelerating search with transferred heuristics. In: Proceedings of the 2007 International Conference on Automated Planning and Scheduling Workshop on Artificial Intelligence Planning and Learning. 2007

[34] Yang Q. Three challenges in data mining. Frontiers of Computer Science in China, 2010, 4: 324−333

[35] Liu Y. A Review about transfer learning methods and applications. In: Proceedings of the 2011 International Conference on Information and Network Technology. 2011, 4: 7−11

[36] Pan W, Zhong E, Yang Q. Transfer learning for text mining. Mining Text Data, 2012, 223−257

[37] Xie Y F, Su S Z, Li S Z. A pedestrian classification method based on transfer learning. In: Proceedings of the 2010 International Conference on Image Analysis and Signal Processing. 2010, 420−425

[38] Rodner E, Denzler J. Learning with few examples for binary and multiclass classification using regularization of randomized trees. Pattern Recognition Letters, 2011, 32: 244−251

[39] Estévez J I, Toledo P A, Alayón S. Using an induced relational decision tree for rule injection in a learning classifier system. In: Proceedings of the the IEEE Congress on Evolutionary Computation New Orleans. 2011, 647−754

[40] Boström H. Induction of recursive transfer tules. In: Cussens J, Džezroski S, eds. Learning Language in Logic. Berlin: Springer, 2000, 1925: 369−450

[41] Reid M D. DEFT guessing: using inductive transfer to improve rule evaluation from limited data. Dissertation for the Doctoral Degree of Philosophy. Sydney: University of New South Wales, 2007

[42] Lee J W, Giraud-Carrier C. Transfer learning in decision trees. In: Proceedings of the 2007 International Joint Conference on Neural Networks. 2007, 726−731

[43] Lu B, Wang X, Utiyama M. Incorporating prior knowledge into learning by dividing training data. Frontiers of Computer Science in China, 2009, 3: 109−122

[44] Ganchev P, Malehorn D, Bigbee W L, Gopalakrishnan V. Transfer learning of classification rules for biomarker discovery and verification from molecular profiling studies. Journal of Biomedical Informatics, 2011, 44 (Suppl 1): S17−S23

[45] Schlimmer J C, Fisher D. A case study of incremental concept induction. In: Proceedings of the 5th National Conference on Artificial Intelligence. 1986, 496−501

[46] Utgoff P E. ID5: an incremental ID3. In: Proceedings of the 5th International Conference on Machine Learning. 1988, 107−120

[47] Utgoff P E. Incremental induction of decision trees. Machine Learning, 1989, 4: 161−186

[48] Michalski R S, Mozetic I, Hong J, Lavrac N. The multi-purpose incremental learning system AQ15 and its testing application to three medical domains. In: Proceedings of the 5th National Conference on Artificial Intelligence. 1986, 1041−1045

[49] ElGibreen H, Aksoy M S. RULES family: where does it stand in inductive learning? In: Proceedings of the 8th International Conference on Computer Engineering and Applications. 2014, 177−186

[50] Maclin R, Opitz D W. Popular ensemble methods: an empirical study. Journal of Artificial Intelligence Research, 1999, 11: 169−198

[51] Hsu K W, Srivastava J. Improving bagging performance through multialgorithm ensembles. Frontiers of Computer Science in China, 2012, 6: 498−512

[52] Srihari S, Yang X, Ball G. Offline chinese handwriting recognition: an assessment of current technology. Frontiers of Computer Science in China, 2007, 1: 137−155

[53] Dai R, Liu C, Xiao B. Chinese character recognition: history, status and prospects. Frontiers of Computer Science in China, 2007, 1: 126−136

[54] Pham D T, Afify A A. Machine-learning techniques and their applications in manufacturing. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 2005, 219(5): 395−412

[55] Lehre P, Yao X. Runtime analysis of search heuristics on software engineering problems. Frontiers of Computer Science in China, 2009, 3: 64−72

[56] Jiang L. Learning random forests for ranking. Frontiers of Computer Science in China, 2011, 5: 79−86

[57] Bigot S. A study of specialisation and classification heuristics used in covering algorithms. In: Proceedings of the 5th Virtual Conference on Innovative Production Machines and Systems. 2009

[58] Janssen F, Fürnkranz J. On the quest for optimal rule learning heuristics. Machine Learning, 2010, 78: 343−379

[59] Lee C. Generating classification rules from databases. In: Proceedings of the 9th International Conference on Applications of Artificial Intelligence in Engineering. 1994: 205−212

[60] Mitchell T M. Machine Learning. New York: McGraw-Hill, 1997

[61] Fayyad U M, Irani K B. Multi-interval discretization of continuousvalued attributes for classification learning. In: Proceedings of the 13th International Joint Conference of Artificial Intelligence. 1993

[62] Cai Z. Technical Aspects of Data Mining. Dissertation for the Doctoral degree. Cardiff, UK: University of Wales Cardiff, 2001

[63] Pham D T, Afify A A. Online discretization of continuous-valued attributes in rule induction. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2005: 829−842

[64] Luengo J, García S, Herrera F. On the choice of the best imputation methods for missing values considering three groups of classification methods. Knowledge and Information Systems, 2012, 32: 77−108

[65] Deogun J, Spaulding W, Shuart B, Li D. Towards missing data imputation: a study of fuzzy k-means clustering method. In: Proceedings of the 4th International Conference of Rough Sets and Current Trends in Computing. 2004, 573−579

[66] Cohen W W. Fast effective rule induction. In: Proceedings of the 12th International Conference on Machine Learning. 1995, 115−123

[67] Michalski R. On the quasi-minimal solution of the general covering problem. In: Proceedings of the 5th International Symposium on In formation Processing. 1969, 128−128

[68] Guan S U, Zhu F. An incremental approach to genetic algorithms based classification. IEEE Transactions on Systems, Man, and Cybernetics, Part B, 2005, 35: 227−239, 2005

[69] Quinlan J R. C4.5: Programs for Machine Learning. San Francisco: Morgan Kaufmann, 1993

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg