As a huge amount of tweets become available online, it has become an opportunity and a challenge to extract useful information from tweets for various purposes. This chapter proposes a novel way to extract topical structure from a large set of tweets and generate a usable summarization along with related topical keywords. Our system covers the full span of the topical analytics of tweets starting with collecting the tweets, processing and preparing them for text analysis, forming clusters of relevant words, and generating visual summaries of most relevant keywords along with their topical context. We evaluate our system by conducting a user study and the results suggest that users are able to detect relevant information and infer relationships between keywords better with our summarization method than they do with the commonly used word cloud visualizations. © 2014 Springer International Publishing Switzerland.
CITATION STYLE
Kas, M., & Suh, B. (2014). Computational framework for generating visual summaries of topical clusters in twitter streams. Studies in Computational Intelligence, 526, 173–199. https://doi.org/10.1007/978-3-319-02993-1_9
Mendeley helps you to discover research relevant for your work.