A Comparison between a KNN Based Approach and a PNN Algorithm for a Multi-label Classification Problem

Techniques for categorization and clustering, range from support vector machines, neural networks to Bayesian inference and algebraic methods. The k-Nearest Neighbor Algorithm (KNN) is a popular example of the latter class of these algorithms. Recently, a slight modification of it has been proposed so that the Multi-Label k-Nearest Neighbor Algorithm (ML-KNN) can deal better with multi-label classification problems. In this paper we are interested in automatic text categorization, which are becoming more and more important as the amount of text in electronic format grows and the access to it becomes more necessary and widespread. We proposed a Probabilistic Neural Network Algorithm (PNN) tailored to also deal with multi-label classification problems, and compared it against the ML-KNN algorithm. Our implementation surpass the ML-KNN algorithm in four metrics typically used in the literature for multi-label categorization problems.