Prediction based Policy setting by finding significance of Attributes from the Ontological Framework in Agricultural domain

In this paper, a methodology has been proposed to tackle the problem of identifying and processing the information conveyed by a huge number of research papers and policy documents, specifically in Agricultural domain. The methodology used in this work includes a series of natural language processing techniques. Pre-processing techniques such as normalizing text to lowercase, stop word removal, tokenization are performed, followed by sentiment analysis for which an algorithm based on the ontological framework has been written. This algorithm works on same concept as the page rank algorithm. Word clouds have been used for analysis. A histogram representing the frequency of occurrence of words in each column is made to identify which domain is being emphasized on by the document.