Personal tools
You are here: Home Research Theses Extracting Opinions from Blog Comments

Extracting Opinions from Blog Comments

Extracting Opinions from Blog Comments: Analysis, Design and Applications - By Preethi Raghavan, Sponsored by John Glenn School of Public Affairs, the Ohio State University

The growth and popularity of online interactive media such as blogs, discussion forums etc. present many challenges and opportunities as organizations and individuals attempt to analyze and comprehend the collective opinion of others using information technology. This thesis will address the need for an open source sense-respond cyber infrastructure framework that an organization could apply to extract and analyze the “voice of the customer”, from blog comments, helping decision makers develop strategic, tactical and operational planning initiatives. It allows organizations to achieve business process improvement and generate inputs for modeling functional and non-functional requirements for IT and other initiatives. The central hypothesis is that knowledge can be obtained by mining opinion and sentiment from comments on blogs and that helps generate business intelligence for an organization. This thesis makes the following contributions. First, it describes the process used to extract and analyze blog comments including the algorithms used for cluster analysis and summarization and the comment analysis workflow. This is followed by details of the opinion mining tool interface design using cognitive engineering principles. It then presents the sense-respond framework and explains how it can be used to draw business intelligence from mined opinion. The tool serves as a proof of concept to validate the sense portion of the framework. Finally, the experiments done on various comment datasets are described and the results are presented. Click here for poster

Student: Preethi Raghavan

Sponsored by John Glenn School of Public Affairs, the Ohio State University

Document Actions
« September 2017 »
September
MoTuWeThFrSaSu
123
45678910
11121314151617
18192021222324
252627282930