Comparing Time Series Sentiment Analysis Results of Social Media Big Data and Google Trends
Project Team: Meltem Özturan, Mustafa Coşkun
Social media can be defined as the main platform for people to express their feelings today. It is also clear that the second platform for detecting the behaviors and feelings of society would be the search engines. In this perspective, in this project, we will try to compare the sentiment analysis results of social media big data and search engine trends data.
First, a sample of Twitter users will be accessed from trending topic search results. Those users will be filtered with related filtering methods to crate sample of the study. Then, tweets of those users will be collected with Twitter APIs. The resultant tweets will be analyzed with sentiment analysis software (SentiStrenght V2).
Second, the sentiment results of the tweets will be arranged in weekly base, because Google trend correlation analysis (https://www.google.com/trends/correlate) can only be done in weekly base. Afterwards, the sentiment data will be applied to Google trend correlate progress and the trending topics with related to the sentiment results will be evaluated.