Exploring Topic Models on Short Texts: A Case Study with Crisis Data

In recent years, social media platforms like Twitter and Facebook have become one of the crucial sources of information for a wide spectrum of users. As a result, these platforms have also become great resources to provide support for emergency management. During any crisis, it is necessary to sieve through a huge amount of social media texts within a short span of time to extract meaningful information from them. Extraction of topic keywords from these unstructured social media texts play a significant role in building any application for emergency management. Topic models have the ability to discover latent topics and latent feature representations of words in a document collection and can be used for this purpose. The main aim of this paper is twofold: to explore topic model implementations and look at its effectiveness on short messages (as short as 140 characters); and to perform an exploratory data analysis on short texts related to crises collected from Twitter, and look at different visualizations to understand the commonality and differences between topics and different crisis related data.