Twitter sentiment analysis using Machine Learning and Hadoop: A comparative study
Anshu Bhasin, Suman R. Das · 2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC) · 2021
Nowadays Big Data is one of the most commonly used terms. As the name suggest it is an enormous amount of data which is very difficult to handle. Twitter is a very famous social networking sites and a very good source of Big Data. The main objective of Twitter Sentiment analysis is to find a way to discover what people like, what they prefer or how they feel about a particular topic or product. We can find out sentiment analysis in web mining, data mining; it is a part of Natural Language processing. This paper adds to arrangement of tweets into one or the other positive or negative utilizing Machine Learning procedures like Naïve Bayes, Logistic regression model, Support vector machine and an another platform Hadoop. We observe that the accuracy of the machine learning methods varies from one dataset to another. We also observe that Apache Spark is faster and more efficient than HDFS for dynamic dataset.