Fast Multi-label Text Classification Algorithm Based On Cost Sensitive

During the past decade, with the exponential growth of information on the Internet, processing of massive mounts of data has become the key challenge. Text classification as an important intelligent process technology has been widely applied in the applications such as information filter, information retrieval and database technology. Various feasible methods are proposed to solve the above issues. However, how to retrieve the massive text information is still an open problem.In this article, we aim to provide an efficient solution to the multi-label text classification, with the combination of text preprocessing, text transformation and feature selection. The fast mul-label text classification based on cost sensitive is proposed for efficient text classification and retrieval, which utlizes the locality sensitive hashing method to fast nearest neighbor search and use the cost sensitive technology to improve the accuracy of the classification. Extensive simulation are conduceted in the real world data, and the results show that our proposed method effectiveness and superiority.