Abstract:
Recent advancements in Web have enabled the wide scale participation in content, which has changed the way we communicate and access information. Traditionally people were subject to accessing news from main stream media such as newspapers and televisions. With the advent of participatory web, citizen journalism has emerged, which manifests itself on the Web. The most recent form of Web publication are microblogs. Microblogs are similar to blogs, which are timestamped posts that are most typically consumed through subscription. Unlike conventional blogs, however, microblogs differ in their tiny size (140 characters) and the frequency of posting. Microblogs are used for a variety of reasons. Social interaction, information gathering, information sharing, marketing, and spam are among the key uses. Thought and feelings regarding global events are sure to reflect on microblogs. The massive quantity of posts make these platforms interesting to explore with respect to news information sharing behavior. At the same time, the massive quantity also makes it a challenge to identify posts that are informative. This thesis proposes an approach for identifying news posts. This approach is implemented in order to fetch and analyze such tweets to study temporal and quantitative properties of such posts. • How do main stream media use microblogs? • Do individuals share news If so, how? • What are the temporal and quantitative properties? Twitter, the most popular microblogging system (at the time of the writing of this thesis) was used as microblogging system. The posts (called tweets) were filtered to identify news tweets, both for news events as well as individuals. The results based on 60 users and various news events are presented. In this thesis, a news pattern to identify news contributions was introduced. Twitter was chosen as a microblog source.A number of tweets related to different global events and individual microblog user posts were examined.
