Effective Pattern Discovery for Text Mining using Neural Network Approach
It is observed that the text mining using the pattern discovery normally uses only the text material in normal fonts i.e. it does not consider the bold, underlined or italic or even the larger fonts as the key text pattern for text mining. This creates problem many a times when the key words are extracted from the document by the algorithm itself. In that case, important keywords are left from the main stream of text patterns. In the proposed paper, patterns are mined in both positive and negative feedback.