Modeling Recommendation System for Real Time Analysis of Social Media Dynamics

With the increasing popularity of twitter, Sentiment analysis of data from twitter has become a research trend. With the help of Twitter API, large number of tweets can be retrieved in real time related to our interest for the analysis. Millions of tweets are posted daily which contain opinions of users around the world. The aim of this project is to develop a desktop application which present users with tweets they may have an interest in. This model analyzes the most used keyword and most mentioned username from the user timeline and henceforth recommend recent tweets with the same keyword from that user. The proposed model assists user in finding relevant tweets related to a particular keyword or hashtag which are categorized into positive and negative using sentimental analysis. The model also deals with different visualization techniques to illustrate the analysis of the most used keyword and mentioned username with the help of various R packages.