Data with Shifting Concept Classification Using Simulated Recurrence

One of the serious problems of modern pattern recognition is concept drift i.e., model changing during exploitation of a given classifier. The paper proposes how to adapt a single classifier system to the new model without the knowledge of correct classes. The proposed simulated concept recurrence is implemented in the non-recurring concept shift scenario without the drift detection mechanism. We assume that the model could change slightly, what allows us to predict a set of possible models. Quality of the proposed algorithm was estimated on the basis of computer experiment which was carried out on the benchmark dataset.

[1]  Gerhard Widmer,et al.  Learning in the Presence of Concept Drift and Hidden Contexts , 1996, Machine Learning.

[2]  Michal Wozniak,et al.  Artificial Recurrence for Classification of Streaming Data with Concept Shift , 2011, ICAIS.

[3]  Abdelhamid Bouchachia Adaptive ambient intelligence and smart environments [invited talk] , 2011 .

[4]  Koichiro Yamauchi,et al.  Learning, detecting, understanding, and predicting concept changes , 2009, 2009 International Joint Conference on Neural Networks.

[5]  Terry Windeatt,et al.  Diversity measures for multiple classifier system analysis and design , 2004, Inf. Fusion.

[6]  Philip S. Yu,et al.  Mining concept-drifting data streams using ensemble classifiers , 2003, KDD '03.

[7]  Ludmila I. Kuncheva,et al.  Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy , 2003, Machine Learning.

[8]  Thorsten Joachims,et al.  Detecting Concept Drift with Support Vector Machines , 2000, ICML.

[9]  A. Campbell,et al.  Progress in Artificial Intelligence , 1995, Lecture Notes in Computer Science.

[10]  Jun-Min Chen,et al.  A new measure of classifier diversity in multiple classifier system , 2008, 2008 International Conference on Machine Learning and Cybernetics.

[11]  Michal Wozniak,et al.  Algorithm of designing compound recognition system on the basis of combining classifiers with simultaneous splitting feature space into competence areas , 2009, Pattern Analysis and Applications.

[12]  Grigorios Tsoumakas,et al.  An Ensemble of Classifiers for coping with Recurring Contexts in Data Streams , 2008, ECAI.

[13]  Mykola Pechenizkiy,et al.  Online mass flow prediction in CFB boilers with explicit detection of sudden concept drift , 2010, SKDD.

[14]  Janez Demsar,et al.  Statistical Comparisons of Classifiers over Multiple Data Sets , 2006, J. Mach. Learn. Res..

[15]  Thomas G. Dietterich Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[16]  Xindong Wu,et al.  Mining Recurring Concept Drifts with Limited Labeled Streaming Data , 2010, TIST.

[17]  Ludmila I. Kuncheva,et al.  A framework for generating data to simulate changing environments , 2007, Artificial Intelligence and Applications.

[18]  Alessandra Russo,et al.  Advances in Artificial Intelligence – SBIA 2004 , 2004, Lecture Notes in Computer Science.

[19]  Sargur N. Srihari,et al.  Decision Combination in Multiple Classifier Systems , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  João Gama,et al.  Learning with Drift Detection , 2004, SBIA.

[21]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[22]  João Gama,et al.  Tracking Recurring Concepts with Meta-learners , 2009, EPIA.