A Survey of Recommendation Systems in Twitter

摘要

Social data mining is a major research area in this new era of technology. Various popular social networking websites such as Facebook, YouTube, Twitter provide a platform to the people for exchanging information and maintain a connection with the friends, relatives, and other active users. Twitter presents a platform to the active users for expressing their views and opinions on a trending topic by posting 280-character tweets. This feature of Twitter makes it different from other social networking sites. This popular microblogging site has approximately 328 million users and the number of tweets that are generated every day is approximately 500 million. Hence, the amount of information that users receive daily on their timeline is quite large. Recommender Systems have been introduced to solve this major problem of information overload. These systems help users to find useful and interesting information. Information filtering is a major step to provide important and useful tweets to the active users. They may miss out the important information due to overwhelming tweets on their timeline. The paper presents the different approaches and techniques that recommender systems have implemented to recommend the important tweets as well as followees to the users based on their behavior and other important features.