The social media has intensely transformed the way people express facts and opinions. In addition to traditional products reviews at the merchant sites which have very limited data and use, people can now express their views on almost anything on the social media sites. For any organization (public or private), this new paradigm of social media contents is likely to replace the traditional methods i.e. surveys, focus groups or consultancy services in order to find masses public opinions about its products, services or institutions. Because extracting public opinions from social media can save both time and money, which by traditional ways are both time consuming and money exhaustive. Till now no known "aspect level sentiment analysis" research is conducted on social media contents. This research is the initial step for formulating a technique for aspects based sentiment analysis of social media sites especially Twitter. The proposed technique in this research will help companies to build new applications for social media or Big Data sentiment analytics. This research will help them evaluate their customer?s opinions about the products and services from different perspectives and levels.