The Application of Latent Semantic Indexing in the Task of Spam Filtering
Pengming Wang · Journal of Zhengzhou University · 2010
The classification performance of latent semantic indexing(LSI) applied to the task of spam filtering is studied.Comparisons to the simple vector space model(VSM) and to the extremely widespread,de-facto standard for spam filtering,the SpamAssassin system,are summarized.Moreover,a set of purely textual of E-mail messages that are based on standard word-and token-extraction techniques,and a set of application-specific meta features of E-mail messages as extracted by the SpamAssassin system are compared.It is shown that VSM and LSI achieve significantly better classification results than SpamAssassin.