Extracting localized information from a Twitter corpus for flood prevention
This work addresses flood prevention for disaster management by providing methods to process social media data, but it appears incremental as it builds on existing corpus analysis techniques without major breakthroughs.
The paper tackled the problem of extracting localized information from Twitter for flood prevention by analyzing a corpus from tropical storm Harvey, focusing on spatial quality estimation and topical representation strategies for unlabeled tweets.
In this paper, we discuss the collection of a corpus associated to tropical storm Harvey, as well as its analysis from both spatial and topical perspectives. From the spatial perspective, our goal here is to get a first estimation of the quality and precision of the geographical information featured in the collected corpus. From a topical perspective, we discuss the representation of Twitter posts, and strategies to process an initially unlabeled corpus of tweets.