ANALYSIS OF TWITTER DATA USING LOGISTIC REGRESSION CLASSIFICATION BASED MACHINE LEARNING METHOD

Abstract

Web innovation has shown great improvement and there is an enormous extent of information
present in the web for web utilizers and many data is delivered also. Web has turned into a stage for internet
getting the hang of, trading thoughts and imparting insights. Twitter is one of the eminent social networking
site, which examine this work where people reveal their personal information, express their views and
notions, interests, likes and dislikes in the site with or without knowledge. To identify the users opinion
such as positive and negative comments the system required machine-learning techniques. For this purpose,
this work concentrated on novel supervised machine learning based classification technique. To obtain high
classification accuracy, the proposed system introduced with preprocessing and k-means clustering
approach. parameters like F-Measure, accuracy, recall and precision are the most important factors for a
social network analysis system. Our proposed system produced better results when compared with the
existing approaches.