A Novel Approach to Classify Train Reviews Based on Sentiment Analysis and Compare the Probability of Error Rate over Hadoop Architecture

Abstract

Big Data and Sentiment Analysis are drawn big attention in public social data due to complex analysis of large volume of data. This paper presents here sentiment analysis and classification of train data using Hadoop Architecture. Trains are the transportation mode for 40 per cent of the world 's passengers. Trains are unlike buses that get into frequent traffic jams, but even trains can be delayed due to problems with engines, engineering problems, problems with passengers or natural disasters. The passengers face different problems due to train delays.While we cannot avoid any delays, we can forecast the delays by observing the conditions of the train stations, such as weathers. There are also other issues such as how passengers will be treated on the waiting list. Such data sets would render a large volume of data waiting to be analyzed without data incoherence.The Big Knowledge Technology uses all of the information to approximate once and to predict how the delay can occur, so that some could use different trains to travel. In addition, we will use time ticketing schemes on the train ticketing schemes so that the passenger roster will be managed in our future proposal by Spark technology.