PROJECT TITLE :

Discovering Newsworthy Themes From Sequenced Data: A Step Towards Computational Journalism - 2017

ABSTRACT:

Automatic discovery of newsworthy themes from sequenced knowledge will relieve journalists from manually poring over a large quantity of information in order to seek out fascinating news. During this paper, we tend to propose a unique k -Sketch query that aims to search out k placing streaks to best summarize a topic. Our scoring perform takes into consideration streak strikingness and streak coverage at the identical time. We study the k -Sketch question processing in each offline and on-line eventualities, and propose numerous streak-level pruning techniques to search out hanging candidates. Among those candidates, we tend to then develop approximate strategies to discover the k most representative streaks with theoretical bounds. We conduct experiments on four real datasets, and also the results demonstrate the potency and effectiveness of our proposed algorithms: the running time achieves up to 500 times speedup and the quality of the generated summaries is endorsed by the anonymous users from Amazon Mechanical Turk.