Comparative Study of Sentiment Analysis of Type 1 and Type 2 Diabetic Patients Using Apache Flume and Hive

Abstract

The Hadoop ecosystem platform has been used to perform sentiment analysis of the opinion regarding type 1 and type 2 diabetes. This research work concentrates on exploring the public opinion about assumptions, thinking and behavior towards the condition of type 1 and type 2 diabetes. Tweets from twitter have been taken as dataset for sentiment analysis for the research. After conducting a set of repetitive experiments, it has been concluded that people feel more negative about type 2 diabetes. They are more pessimistic about the diet and treatment of type 2 diabetes. Present paper is divided mainly into four sections. The first section, gives an overview of the sentiment analysis and diabetes. The proposed work and methodology have been included in the second section. The third section contains the whole experiment work to support the results. In the fourth section, results regarding opinions about type 1 and type 2 diabetes have been illustrated using bar charts obtained from the experiments followed by concluding remarks, future scope and significance of the research.