The problem the world is facing today is the large amount of data i.e. the data is increasing each and every day in an enormous amount. The growth is so large that the data which has been found over last year i.e. approximately 5 zeta bytes can be seen today within two or three days. The world not only facing the problem of storing of this massive amount of data but also the methods to access and analyse this large amount of data in specified amount of time. So to overcome this problem a concept has been evolved over past few years in technological field i.e. “Big Data”. The most powerful platform responsible for this massive growth of data is “Social Networks”. A large amount of data is generated on Social Networks in a day and that can be structured,unstructured and semi-structured and is one of the most widely used platforms for everyone. There are various implementations to solve the Big Data problems and among them one of the most popular and widely used implementation is using the Hadoop and Map Reduce programming. This paper aims to analyse social media (Twitter,etc.) big data to identify the widespread of certain keywords related to different diseases at various locations such as India,TamilNadu and World. It will be used to identify the number of people who are using these keywords in a social network classified by the location of the people. It is going to fetch social media data set using Social Networks API and data set is stored in HDFS and a Map Reduce program is to be written to obtain the number of times a disease keyword is used by a certain number of people on their geographic location and finally visualize the results using graph. Also,classification techniques such as Naïve Bayesian classification can be applied to the categorical data. © 2016,International Journal of Pharmacy and Technology. All rights reserved.