PREEMINENT FEATURE DISCOVERY AND DOCUMENT CLUSTERING USING TEXT MINING

Abstract

Text document classification and indexing is the major part of document management, where every document
should be identified by its key terms and domain knowledge. Based on the domain knowledge, the documents are classified
into different classes. For document classification there are several approaches were proposed in existing system. But the
existing system is either term based or pattern based. And those systems suffered from polysemy and synonymy problems.
To make a revolution in this challenging issue, the proposed system presents an innovative model for relevance
feature discovery and document classification. It discovers both positive and negative patterns in text documents as higher
level features and deploys them over low-level features (terms). It also detects the most appropriate features based on its
weight and semantic nature and performs the document classification. Using this approach, the document index terms,
patterns and category can be identified easily. In order to perform the above, a hybrid approach is used which contains the
following algorithms. (a). sequential semantic pattern mining algorithm for sequential pattern extraction (b). Semantic
Weighted feature ranking algorithm to rank the higher supported terms in the form of semantic and patterns. (c). a Rule
based ontology concept which helps to classify the documents under various classes and helps to detect the possible index
terms. This helps to reduce the training data collection problems.