LD- SMO Algorithm for Determining Trust in Skewed Social Media Data

Collaborative Web Applications (CWAs) have become a pervasive part of internet. Social media (such as Face book, twitter), topical forums, wikis are all examples of CWAs – that enable a community of end users to interact or cooperate towards a common goal. Some key characteristics of CWAs such as low entry barrier, instant updates, large number of friends, open platform, anonymity make it vulnerable to activities of ill intentions thereby providing a medium for nefarious persons to operate. One such collaborative system is Twitter which has experienced enormous growth in a small amount of time and users do spend respectable amount of their everyday time interacting about various topics with other peers. Twitter has evolved from being a conversation or opinion sharing medium among friends into a platform to share and disseminate information about current events. Events in the real world create a corresponding spur of posts (tweets) on Twitter. In other words it has evolved from a micro blogging service to a major news source. Although a large volume of content is posted on Twitter, not all information is trustworthy or useful in providing information about the event. Gossips, fake news etc. are also a part of genuine news. The main aim of this paper is to tackle the issue of accuracy paradox, a major problem when dealing with social media research, were the data extracted by us was highly imbalanced. This high imbalance in the data was solved by designing an LD-SMO algorithm which achieved an accuracy of ~96% with an equally comparable sensitivity and specificity.