ISSN(Online) : 2319-8753 ISSN (Print) : 2347-6710 International Journal of Innovative Research in Science, Engineering and Technology (An ISO 3297: 2007 Certified Organization) Vol. 5, I ssue 5, May 2016 Copyright to IJIRSET DOI:10.15680/IJIRSET.2016.0505246 8355 Sentiment Analysis using Maximum Entropy Algorithm in Big Data Durgesh Patel 1 , Sakshi Saxena 2 , Toran Verma 3 P.G. Student, Department of Computer Engineering, Rungta College of Engineering & Technology, Bhilai, India 1 Assistant Professor, Department of Computer Engineering, Rungta College of Engineering & Technology, Bhilai, India 2 M. Tech Coordinator, Department of Computer Engineering, Rungta College of Engineering & Technology, Bhilai, India 3 ABSTRACT: As years are passing people are getting more and more attracted towards Social Media Sites and they had started attaching there life with these electronic gadgets. As this things are happening the data being created and shared by millions of people that is increasing per second and these data are not easy to store and manage. As we know that everything in this world has some meaning full existence in this world in that similar way this Big data can also be used for some application and for that the analysis done is called Sentiment Analysis. But Analyzing this large dataset is not so easy , and same way Sentiment Analysis is lacking for Negation Statement case. This problem is reducing the accuracy of Sentiment Analysis which can be verified easily with the proposed methodology in this paper. KEYWORDS: Negation, Sentiment Analysis, Big data I. INTRODUCTION ig Data is the most emerging word during this period of time and it is the most major topic to be concerned for. Social Media has taken over a very large space in human life style that is why Big data is also became very big problem to handle and use it simply[4]. And this data from social media is very important and can be used for various analysis like Sentiment analysis. With the innovation of digital Technologies and smart devices , a huge amount of digital data is being generated everyday. This data is very hard to manage using conventional techniques like Relational Database Management System[7] , these system can not able to store this huge data and impossible to analyze this data fastly and accurately. Due to the failure of conventional technique of storage of data for big Data a new system is invented termed as Hadoop , Hadoop is a open source framework which has ability to store the data and also to analyze the data accurately , precisely and fastly also. The Apache Hadoop software library is a framework that allows for the distributed processing of large data sets across clusters of computers using simple programming models. It is mainly designed to maintain thousands of machine through a single server. Hadoop has various tools which supports different languages for its processing , example, HDFS , Hive , Map-Reduce , Pig , Oozie , flume etc [13]. This paper will mainly need to get in touch with Flume, Hive , HDFS, Pig. Flume supports for the collection of dataset from the developers site of Twitter (i.e http://dev.twitter.com) and hive mainly uses HSQL language which is for the data handling as in tables, views etc. HDFS is mainly for the storage purpose. And in this project Pig is used for making a user defined function for different procedure. As various studies has been conducted in this area but the accuracy of sentiment analysis is not up to the mark till now. And in case of Big Data the accuracy is not good because to handle large amount of data is not an easy task. Our main purpose is not only to store large amount of data but also to analyze it for the benefit of organizations and also for the peoples welfare. B