NebFuzz: Um Novo Algoritmo de Agrupamento Semi-Supervisionado Baseado no Fuzzy C-Means

Valmir Maca rioFrancisco de A. T. de Carvalho

Semi-supervised clustering uses unlabeled data, combined with the labeled data, to improve the algorithm performances. This paper presents a new algorithm for semi-supervised clustering based on Fuzzy C-Means algorithm. The new algorithm was evaluated and compared against two semi-supervised clustering algorithms in the context of learning from partially labeled data. The behavior of the proposed algorithm is discussed and the results are validated using accuracy rate, corrected rand index and a 95% confidence interval. Thus, it was possible to certify the better accuracy performance of the new algorithm when a few labeled data are available.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: