An Enhanced Text Mining Approach using Dynamic Programming

Text mining is a pervasive area of research in Computer Science. This is the process of finding knowledgeable and useful information and patterns from very huge text. It is widely applied in several areas of applications such as information retrievals; computational biology etc. Keyword-based searching has been extensively applied in text dataset considering the keywords as strings. String matching is used in finding all the occurrences of a given pattern P of length m from a given text T of length n, where m n. An occurrence of a pattern inside the text can simply be characterized as "exact" or "approximate". This paper proposes a framework for text mining using a fast bitparallel algorithm for searching exact occurrences of a pattern inside a huge body of text. We evaluate the performance of three algorithms in the literature on different text files and discuss their suitability under different situations.