A Review on Semantic Graph based Text Mining Manju Lata

  • Published 2015


In the universal framework of knowledge discovery, Data Mining techniques are exhaustively devoted to information extraction from structured databases. On the other side, Text Mining techniques are keen to information extraction and retrieval from unstructured textual data. Several models have been introduced for effective text mining. One of the extensively used model is Vector Space Model (VSM) which is based on beg of words approach. Apart from its popularity, it suffers from a range of limitations, as it is based on frequency of the terms in the document, it losses the semantics of the terms in the document which may cause of mislead to several interpretations. To overcome from this, other data models based on the semantics are becoming very well-accepted. One of such data model is Ontology. In this paper we discuss an Ontology based approach to construct semantic graph for representing text document. Semantic graph explores the semantics of the document by considering semantic relations (derived from ontology) among the terms of the document. Once the text document is represented into semantic graph, it can be used for a number of applications of text mining such as Question Answering, Concept Generation, Keyword Extraction, Semantic Indexing, Automatic Text Summarization, Clustering, and Categorization etc. in relevance with text mining. The paper also discusses graph measures to extract linguistic properties of text document. We conclude the paper with the discussion on few state of the art work done in the same direction.

Cite this paper

@inproceedings{2015ARO, title={A Review on Semantic Graph based Text Mining Manju Lata}, author={}, year={2015} }