Linked Data the Challenges for Filtering Social Data Semantic Filtering for Social Data

74 Published by the IEEE Computer Society 1089-7801/16/$33.00 © 2016 IEEE IEEE INTERNET COMPUTING O ne in three Web users looks for medical information on social networks (http://bit. ly/wiredarabspring), and more than 50 percent of users surveyed consume news on social networks (http://bit.ly/pewsnsnews). Twitter and Facebook were prominent platforms used for disseminating information and organizing protests during Arab Springs, Occupy Wall Street, and similar events.1 Social data also plays a critical role in helping with coordination during natural disasters. Social networks have therefore not only changed the landscape for communicating and sharing information — they have also become a major source for users consuming information. The popularity of social networks has led to an increase of user-generated content on the popular platforms. Facebook and Twitter together generate more than 5 billion microblogs per day. Because users consume information from these platforms, the overwhelming amount of content generated brings to mind Herbert Simon’s famous quote: “a wealth of information creates a poverty of attention.”2 In turn, the growth in the volume of content has often drawn criticism of information overload from consumers. As users of the Web, it’s important for us to realize that our dependence on information from these platforms will continue to grow, and hence, so should our focus on working towards making our lives easier in accessing the collected intelligence on these platforms, particularly by addressing the problem of information overload. Researchers have addressed the challenge of information overload by developing information filtering systems that understand a group of users’ interests and deliver relevant content to them. Although these filtering techniques have been adopted for filtering spam in emails and delivering relevant news and articles to interested users, leveraging these techniques on social networks and building an efficient informationfiltering system presents distinct challenges, due to social networks’ unique characteristics. Here, we consider how to address those challenges, using a crowd-sourced platform such as Wikipedia.