Knowledge Discovery from Unstructured Data using Sentiment Analysis

Information in a business organization is constantly growing today in a form of structured and unstructured data. According to Gartner, Inc., (2016), a world’s leading information technology research and advisory company, “by 2018 more than half of large organizations globally will compete using advanced analytics”. One of the trends that influences the rapid development of this market is business analytics improvement through enrichment with new methods and algorithms to extract and process data from new sources, providing unstructured data especially from interaction with customers. Social media and review sites are possible sources of unstructured data, where users can express their opinion about products and services by posting comments. Business organizations may explore users` opinion using sentiment analysis methods for identifyting positive and negative opinion, expressed about their products and services on the internet. Knowledge discovery from users` comments requires structuring unstructured data and then applying text and data mining methods and tools. Business Intelligent tools are used to present the results from analysis in a proper way to discover new knowledge and to support decision making process in the organisation. The present paper is focused on introducing main definitions, methodologies and tools for mining and analyzing unstructured data from users` comments and a methodology for knowledge discovery from unstructured data using sentiment analysis is suggested.