A particle swarm optimizer based extension classifier for biomedical document

With the rapid growth of biomedical document in recent years, it is of great importance to develop methods for the automatic classification of large collections of biomedical documents. In this paper, we propose a novel extension classifier based on particle swarm optimizer for classification of large scale of biomedical documents, in which, the biomedical document is represented by space vector model, the similarity between document is measured by extension distance. In order to improve the performance of classifier, we design a modified particle swarm optimizer and employ it to train the weights of attributes. Unlike the traditional classification methods, the classifier we proposed has less calculating amount and higher precision. We evaluate the proposed method on MEDLINE, and the experiment result shows our method outperforms K-NN, Decision tree, Bayes classifier and SVM. © 2012 Binary Information Press January 2012.