Threat detection in online discussions

Abstract

This paper investigates the effect of various types of linguistic features (lexical, syntactic and semantic) for training classifiers to detect threats of violence in a corpus of YouTube comments. Our results show that combinations of lexical features outperform the use of more complex syntactic and semantic features for this task.

Extracted Key Phrases

9 Figures and Tables

Showing 1-10 of 13 references

Ngram-based recognition of threatening tweets

  • Nelleke Oostdijk, Hans Van Halteren
  • 2013
2 Excerpts

Guidelines for the clear style constituent to dependency conversion

  • D Jinho, Martha Choi, Palmer
  • 2012