Copy the page URI to the clipboard
Cano Basave, Amparo Elizabeth; He, Yulan; Liu, Kang and Zhao, Jun
(2013).
URL: http://lang.cs.tut.ac.jp/ijcnlp2013/
Abstract
Social streams have proven to be the most up-to-date and inclusive information on current events. In this paper we propose a novel probabilistic modelling framework, called violence detection model (VDM), which enables the identification of text containing violent content and extraction of violence-related topics over social media data. The proposed VDM model does not require any labeled corpora for training, instead, it only needs the incorporation of word prior knowledge which captures whether a word indicates violence or not. We propose a novel approach of deriving word prior knowledge using the relative entropy measurement of words based on the intuition that low entropy words are indicative of semantically coherent topics and therefore more informative, while high entropy words indicates words whose usage is more topical diverse and therefore less informative. Our proposed VDM model has been evaluated on the TREC Microblog 2011 dataset to identify topics related to violence. Experimental results show that deriving word priors using our proposed relative entropy method is more effective than the widely-used information gain method. Moreover, VDM gives higher violence classification results and produces more coherent violence-related topics compared to a few competitive baselines.