Venkata Kishore Neppalli, Cornelia Caragea, & Doina Caragea. (2018). Deep Neural Networks versus Naive Bayes Classifiers for Identifying Informative Tweets during Disasters. In Kees Boersma, & Brian Tomaszeski (Eds.), ISCRAM 2018 Conference Proceedings – 15th International Conference on Information Systems for Crisis Response and Management (pp. 677–686). Rochester, NY (USA): Rochester Institute of Technology.
Abstract: In this paper, we focus on understanding the effectiveness of deep neural networks by comparison with the effectiveness of standard classifiers that use carefully engineered features. Specifically, we design various feature sets (based on tweet content, user details and polarity clues) and use these feature sets individually or in various combinations, with Naïve Bayes classifiers. Furthermore, we develop neural models based on Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) with handcrafted architectures. We compare the two types of approaches in the context of identifying informative tweets posted during disasters, and show that the deep neural networks, in particular the CNN networks, are more effective for the task considered.
|
Venkata Kishore Neppalli, Murilo Cerqueira Medeiros, Cornelia Caragea, Doina Caragea, Andrea Tapia, & Shane Halse. (2016). Retweetability Analysis and Prediction during Hurricane Sandy. In A. Tapia, P. Antunes, V.A. Bañuls, K. Moore, & J. Porto (Eds.), ISCRAM 2016 Conference Proceedings ? 13th International Conference on Information Systems for Crisis Response and Management. Rio de Janeiro, Brasil: Federal University of Rio de Janeiro.
Abstract: Twitter is a very important source for obtaining information, especially during events such as natural disasters. Users can spread information in Twitter either by crafting new posts, which are called ?tweets,? or by using retweet mechanism to re-post the previously created tweets. During natural disasters, identifying how likely a tweet is to be highly retweeted is very important since it can help promote the spread of good information in a network such as Twitter, as well as it can help stop the spread of misinformation, when corroborated with approaches that identify trustworthy information or misinformation, respectively. In this paper, we present an analysis on retweeted tweets to determine several aspects affecting retweetability. We then extract features from tweets? content and user account information and perform experiments to develop models that automatically predict the retweetability of a tweet in the context of the Hurricane Sandy.
|