Published Paper Details:

Now-a -days, people have a very free and convenient communication environment of internet where they show active participation to demonstrate what they actually feel about a particular event. May it be a poll result, incident, news, political issues, government schemes or any government decision for the citizens. People demonstrate their views by reacting through various parameters available on the social media. Some make a tweet using hash tags while some other prefer giving their views using statements, updating status or even updating their profiles in order to support a particular cause. But what if the predictors already know how the public is going to react on a particular event or news. For this prediction system help us out, in which there is analysis of the responses from public and then predict the future circumstances. Many applications are developed by using prediction systems. Some example are cancer detection, election poll results and many more. The day-by-day growing data can compromise the performance of the prediction system, because its obvious that growing network of data will require more storage and the system will also consume more time for its processing. In practical applications, maintenance of network storage and calculations will be costly with increasing number of nodes in the network. For this effective strategy for reducing processing time must be introduced. To reduce this processing time introducing parallelism concept can help.