1 Various Approaches in Text Pre - processing

Text mining, as an increasingly important field of research in Knowledge Discovery in Data (KDD), concentrates on discovering hidden patterns, rules, regularities and trends from textual data, such as natural language speech or web documents. The structure of textual data is considered implicit, which is different from the structured data that stored in databases. The various natures of textual data and the data in databases cause the major difference between text mining and the traditional data mining (the main stage of KDD). However, it is doable to obtain the knowledge from texts by employing the techniques in data mining if we can rationally explicate the textual data to be structured or semi-structured. In this paper, we summarize the various ways of text preprocessing. Our work is presented with the aim of supporting future work in text mining research.