Information Extraction from Large Text Corpus

Abstract

This project presents a method for automatic extraction of the hyponym-hypernym relations from the text data. Many researchers are working to develop and implement automatic information extraction approach using pre-encoded knowledge and patterns. These are semi-supervised machine learning approaches, such approaches are not useful when user wants to extract more relations or discover any new pattern. Another risk with such system is that, if predefined pattern fails to produce new pattern then the operations depends on this pattern will fail. This project focuses on converting the semi-supervised machine learning approach into unsupervised machine learning approach for fully automatic extracting information from text. The objectives of this study are, avoid pre-encoded pattern for more efficiency and define a method for automatically extracting useful relationships and patterns using an unsupervised machine learning approach. The proposed approach is tested with standard datasets as well as real time research article. The unsupervised machine learning approach gives the better result than semi-supervised machine learning approach in term of information extraction.