WoSIT: A Word Sense Induction Toolkit for Search Result Clustering and Diversification

Abstract

In this demonstration we present WoSIT, an API for Word Sense Induction (WSI) algorithms. The toolkit provides implementations of existing graph-based WSI algorithms, but can also be extended with new algorithms. The main mission of WoSIT is to provide a framework for the extrinsic evaluation of WSI algorithms, also within end-user applications such as Web search result clustering and diversification.

Analysis of Online News Coverage on Earthquakes Through Text Mining

“…There are computational challenges and limitations when exploiting the text mining tools (e.g., Radinsky et al, 2012;Asghar et al, 2014), particularly problems when extracting information from unstructured sources such as websites (Vannella et al, 2014). One challenging problem arises from the use of part of speech taggers, including incorrect tagging of words, vague classification of entities (Asghar et al, 2014;Jurafsky and Martin, 2014), language-related issues (Pinto et al, 2016), extraction of temporal expressions (Mani and Wilson, 2000;Kisilevich et al, 2010;Bögel et al, 2014;Derczynski and Gaizauskas, 2015) and geographical locations (Kisilevich et al, 2010;Piskorski and Atkinson, 2011) as discussed by Gupta (2016).…”