Pre Processing of Twitter's Data for Opinion Mining in Political Context

In the wake of political activism among youth in particular and the whole population in general, everyone is not only eager to share their political orientation but equally curious regarding the voice of the masses. As a known notion, the perfect orifice to this emerging need of political activism can be found on social media platforms, from where the numerous aspects of public opinion can be captured easily. These sites have begun to have a large impact on how people think and act. It is a known phenomenon that public opinion is the largest indicator of success and failure of political parties and is a direct reflection of the party's reign. Where increased sharing of public feedback has increased awareness and promoted accountability, it has also created chaos and confusion for many. Using Twitter, the most popular micro blogging platform, this paper aims to give a method to ease and smooth the task of opinion mining with the help of linguistic analysis and opinion classifiers, which will together determine positive, negative and neutral sentiments for the political parties of Pakistan. A method is provided which pre-processes the raw data of twitter and comparison of two classification techniques to classify this data. That will aspire to capture a snapshot of current political scenario to promote the spirit of accountability, self-analysis and improvement in among Pakistani politicians. Moreover, with this we aim to give general public an important consolidated voice in the realm of politics. (C) 2016 Published by Elsevier B.V.