Emerald Group Publishing Limited
Article Type: Guest editorial From: International Journal of Web Information Systems, Volume 10, Issue 4
Social media has played major role as a communication tool and news dissemination tool together with the traditional news sources such as newspaper, radio and television. The abundance of information available in social media inspires many to explore the possibility to analyze and exploit knowledge that can be derived from it. In this special issue on social media analysis, we present five different papers that show different applications of social media analysis, together with its unique approach or technique.
The first paper titled “Opinion-driven communities’ detection” looks at the use of document analysis technique to perform opinion mining. The “opinion” is not only being detected in the passages of text but also is used to automatically detect a community. A community is defined as a group of people who expressed similar opinions. Based on the same technique, the paper also suggests that similar or contradiction in opinions can be detected quickly during the analysis.
Unlike the first paper that uses documents as the source for data analysis, the other four papers look at micro-blog as the source of data analysis. The four papers aim at different types of applications and analysis. Ranging from personalized recommendation, identifying “real life” tweet, communication model in the crisis situation and identifying tweeter’ s users behavior.
The paper titled “Evaluating Credibility of Interest Reflection on Twitter” presents a technique to measure one’s interest based on the analysis of Web pages that are visited as a result of recommendation by tweets. The technique provides a context analysis to judge the credibility of the interest that has been reflected in the tweeted Web pages. The judgment provides a clue to whether there is a real interest of the user on the tweeted pages or it is only something that the user has some curiosity about the contents.
Another user behavior analysis, although of different kind, is presented by the paper “Behavior Analysis Methods for Twitter Users based on Transitions in Posting Activities”. In this paper, the authors look at the detection of whether a user is a long-term user or a short-term user. The detection is performed based on the observation of the transition in posting activities. By looking at the posting activities across time, the authors suggest that detection of user’s status is possible.
The paper “Two Phase Estimation Method for Multi-classifying Real Life Tweets” analyses the content of tweets to label the content into some categories that the authors called “real life” tweets. The real life tweets are defined as tweets that help user to perform live activity, such as a tweet containing “a train is running late”. The main contribution of the paper is by detecting and labeling two real life labels rather than a single label.
In many crisis situations, such as the Japan’s earthquake, social media has been used as an important mean of communication. The experience of many users during the Japan earthquake’s crisis is captured by the paper “How Do Rumors Spread During a Crisis?: Analysis of Rumor Expansion and Disaffirmation on Twitter after 3.11 in Japan”. Several studies have performed analysis of social media during disaster; however, this paper presents something unique that it studies how rumors emerge and how the disaffirmation of rumors can be detected. The authors also compared the spreading behavior of rumors in crisis and non-crisis situation.
The amount of social media for analysis will only be increasing, and the types of analysis performed will only be limited by our creative imagination. This special issue opens the door for that possibility by providing some examples on how the social media can be used for real life applications.
Maria Indrawan-Santiago, Guest Editor, Faculty of Information Technology, Monash University, Melbourne, Australia