Learning about the Learning Process

This work addresses the problem of mining data stream generated in dynamic environments where the distribution underlying the observations may change over time. We present a system that monitors the evolution of the learning process. The system is able to self-diagnosis degradations of this process, using change detection mechanisms, and self-repairs the decision models. The system uses meta-learning techniques that characterize the domain of applicability of previously learned models. The meta-learns can detect re-occurrence of contexts, using unlabeled examples, and take pro-active actions by activating previously learned models.

[1]  Raj K. Bhatnagar,et al.  Tracking recurrent concept drift in streaming data using ensemble classifiers , 2007, ICMLA 2007.

[2]  Edsger W. Dijkstra,et al.  Self-stabilizing systems in spite of distributed control , 1974, CACM.

[3]  Julio Ortega.  Exploiting Multiple Existing Models and Learning Algorithms , 2007 .

[4]  William Nick Street,et al.  A streaming ensemble algorithm (SEA) for large-scale classification , 2001, KDD '01.

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

[6]  Eugene L. Grant,et al.  Statistical Quality Control , 1946 .

[7]  Mihai Lazarescu,et al.  A Multi-Resolution Learning Approach to Tracking Concept Drift and Recurrent Concepts , 2005, PRIS.

[8]  Xindong Wu,et al.  Mining in Anticipation for Concept Change: Proactive-Reactive Prediction in Data Streams , 2006, Data Mining and Knowledge Discovery.

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

[10]  Johannes Fürnkranz,et al.  An Evaluation of Grading Classifiers , 2001, IDA.

[11]  Shlomo Argamon,et al.  Arbitrating Among Competing Classifiers Using Learned Referees , 2001, Knowledge and Information Systems.

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

[13]  Frank Kirchner,et al.  Performance evaluation of EANT in the robocup keepaway benchmark , 2007, ICMLA 2007.

[14]  Claude Sammut,et al.  Extracting Hidden Context , 1998, Machine Learning.

[15]  Grigorios Tsoumakas,et al.  Tracking recurring contexts using ensemble classifiers: an application to email filtering , 2009, Knowledge and Information Systems.

[16]  Ralf Klinkenberg,et al.  Learning drifting concepts: Example selection vs. example weighting , 2004, Intell. Data Anal..

[17]  Gerhard Widmer,et al.  Tracking Context Changes through Meta-Learning , 1997, Machine Learning.

[18]  David J. Hand,et al.  Advances in intelligent data analysis , 2000 .

[19]  Andreas S. Rath,et al.  Analysis of machine learning techniques for context extraction , 2008, 2008 Third International Conference on Digital Information Management.

[20]  Peter D. Turney.  The Management of Context-Sensitive Features: A Review of Strategies , 2002, ArXiv.