Pruning training corpus to speedup text classification

Abstract

With the rapid growth of online text information, efficient text classification has become one of the key techniques for organizing and processing text repositories. In this paper, an efficient text classification approach was proposed based on pruning training-corpus. By using the proposed approach, noisy and superfluous documents in training corpuses can be cut off drastically, which leads to substantial classification efficiency improvement. Effective algorithm for training corpus pruning is proposed. Experiments over the commonly used Reuters benchmark are carried out, which validates the effectiveness and efficiency of the proposed approach.