Early Prediction of Public Reactions to News Events Using Microblogs

Abstract

Microblog environments like Twitter are increasingly becoming more important to leverage people's opinion on public events. We aim to predict future public reactions to news events by exploiting related tweets. We define public reactions in terms of their dimension and direction. Our system collects and preprocesses tweets, creates an inverted index to search tweets efficiently, filters them with various methods according to news events; and then uses temporal, spatial and textual features to model predictive classifiers. We also create a public-reaction dataset, BilPredict-2017, which includes several events including terrorist attacks in Turkey from 2015 to 2017. We plan to model ensemble classifiers, and evaluate the success of our system on BilPredict-2017.

Event-related microblog retrieval in Turkish

“…Although information retrieval is studied for ad-hoc queries in Turkish document collections [7,8], there are limited studies in terms of event-related microblog retrieval in Turkish. Tweets are searched with respect to news events by using inverted index and boolean search [37]. To the best of our knowledge, this is the first study on event-related microblog retrieval in Turkish that examines query and tweet encoding methods, including FastText and BERT, as well as query selection and text preprocessing.…”