Sentiment analysis study with an emphasis on the integration of different statement polarities and the evaluation of the resulting sentiments

ECU Author/Contributor (non-ECU co-authors, if there are any, appear on document)

Shawn Braddy (Creator)

Institution

East Carolina University (ECU )

Web Site: http://www.ecu.edu/lib/

Abstract: Social media has become an integral part of today's society and has continued to grow tremendously throughout the world. People communicate through social media constantly and social media has become a pivotal place for people to gather information related to things that they are interested in. It is also a place for people to express their views to friends , family and followers with just one click of a button immediately. Naturally in society there are a wide range of views encompassing an unlimited amount of subjects. Some of these views can be deemed negative , positive , neutral or maybe even other. My research will attempt to identify the sentimental impact that different views in the form of social interactions online , can have on the people who are watching , listening and reading them. Furthermore , I will research how the integration of neutral and positive statements can stimulate positive and productive conversation amongst a community on a small scale as well as a large scale. Analysis of this will be quantified by changes in the overall reaction of the various communities time following intervention. The main contribution of this thesis is to show a correlation between statements with positive or neutral sentiment and a trend towards positive or neutral responses after moderation. Events of large discussion or interest will be evaluated to ensure a significant amount of data is available for analysis. In order to integrate the social aspect into the equation , software will be explored and evaluated or built and then utilized to connect participants and extract the appropriate metrics.

Additional Information

Publication

Thesis

Language: English

Date: 2016

Keywords

Sentiment analysis

Subjects

Email this document to