Proposal for a Unified Methodology for Evaluating Supervised and Non-supervised Classification Algorithms
There is presently no unified methodology that allows the evaluation of supervised and non-supervised classification algorithms. Supervised problems are evaluated through Quality Functions that require a previously known solution for the problem, while non-supervised problems are evaluated through several Structural Indexes that do not evaluate the classification algorithm by using the same pattern similarity criteria embedded in the classification algorithm. In both cases, a lot of useful information remains hidden or is not considered by the evaluation method, such as the quality of the supervision sample or the structural change generated by the classification algorithm on the sample. This paper proposes a unified methodology to evaluate classification problems of both kinds, that offers the possibility of making comparative evaluations and yields a larger amount of information to the evaluator about the quality of the initial sample, when it exists, and regarding the change produced by the classification algorithm.
KeywordsClassification Problem Classification Algorithm Structural Index Quality Function Final Covering
- 4.Arbel, R., Rokach, L.: Classifier evaluation under limited resources. Pattern Recognition Letters, Available online (2006)Google Scholar
- 5.Martínez-Trinidad, J.F., Guzman-Arenas, A.: The logical combinatorial approach to Pattern Recognition an overview through selected works. Pattern Recognition 34(4), 1–11 (2001)Google Scholar
- 10.Godoy-Calderón, S., Lazo-Cortés, M., Martínez-Trinidad, J.F.: A non-classical view of Coverings and its implications in the formalization of Pattern Recognition problems. WSEAS Transactions on Mathematics 2(1-2), 60–66 (2003)Google Scholar