Author(s): Dhanasekaran K; Rajeswari R

Text mining discovers unseen patterns from textual data sources. But these discoveries are useless unless they contribute valuable knowledge for users who make strategic decisions. Confronting this issue can lead to knowledge discovery from texts, a complicated activity that involves both discovering unseen knowledge and evaluating this potentially valuable knowledge. Information Extraction can benefit from techniques that are useful in data mining or knowledge discovery. However, we can’t easily apply data mining techniques to text data for text mining because they assume a complex structure in the source content. Therefore there is a need to use new representations for text data. In many like applications, we can use more structured representations than just keywords to perform analysis to uncover unseen patterns. Early research on such an approach was based on seminal work on exploratory analysis of article titles stored in the Medline medical database. Other approaches have exploited these ideas by combining more elaborated information extraction patterns and general lexical resources such as WordNet or specific concept resources such as thesauri. Another approach, relying on Extracted patterns, uses linguistic resources such as WordNet to assist the discovery and evaluation of patterns to extract basic information from general documents. This paper presents an efficient extraction through classification model to support efficient retrieval and data processing applications. In this paper we discuss various approaches related to text mining, explore issues faced in various research.