论文信息 - Combination of Task Description Strategies and Case Base Properties for Meta-Learning

Combination of Task Description Strategies and Case Base Properties for Meta-Learning

Describing a learning task is crucial, not only for metalearning but also to gain insight in this learning task. The paper evaluates the performance of a recent method for assessing quality standards for case bases when used for a supervised meta-learning. Empirical results on real-world data show this approach in combination with others as a promising one.

Ioannis Iglezakis | Christian Köpf

[1] J. R. Quinlan. Induction of decision trees , 2004, Machine Learning.

[2] Hilan Bensusan,et al. Meta-Learning by Landmarking Various Learning Algorithms , 2000, ICML.

[3] Richard O. Duda,et al. Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[4] Hilan Bensusan,et al. Casa Batlo is in Passeig de Gracia or landmarking the expertise space , 2000 .

[5] Hilan Bensusan,et al. Automatic bias learning: an inquiry into the inductive basis of induction , 1999 .

[6] Aiko M. Hormann,et al. Programs for Machine Learning. Part I , 1962, Inf. Control..

[7] William W. Cohen. Fast Effective Rule Induction , 1995, ICML.

[8] Rudi Studer,et al. AST: Support for Algorithm Selection with a CBR Approach , 1999, PKDD.

[9] João Gama,et al. Discriminant Trees , 1999, ICML.

[10] Hilan Bensusan,et al. Estimating the Predictive Accuracy of a Classifier , 2001, ECML.

[11] Catherine Blake,et al. UCI Repository of machine learning databases , 1998 .

[12] David J. Spiegelhalter,et al. Machine Learning, Neural and Statistical Classification , 2009 .

[13] Thomas Roth-Berghofer,et al. On Quality Measures for Case Base Maintenance , 2000, EWCBR.

[14] Thomas Reinartz,et al. Relations between Customer Requirements, Performance Measures, and General Case Properties for Case Base Maintenance , 2002, ECCBR.