Extracting Criminal-Related Events from Arabic Tweets

Abstract

Recently, Twitter as one of social networks has been considered as a rich source of spatio-temporal information and significant revenue for mining data. Event detection from tweets can help to predict more serious real-world events. Such as: criminal events, natural hazards, and the spread of epidemics. Etc. This paper deals with event-based extraction for criminal incidents from Arabic tweets. It presents a framework that supports automated extraction of spatial and temporal information from tweets. The proposed approach is based on combining various indicators, including the names of places and temporal expressions that appear in the tweet message, related tweeting time, and additional locations from the user's profile. The effectiveness of the system was evaluated in term of recall, precision and f-measure.

A Parallel Corpus-Based Approach to the Crime Event Extraction for Low-Resource Languages

“…The main reason for the absence of a unified standard approach for EE lies in the fact that ML approaches need large, semantically annotated corpora but a pattern-based event extraction approach is a time-consuming and laborintensive task that must involve a lot of domains. Therefore, the additional methodological contribution of our research is the enhancement of pattern-based event extraction method [14], [16], [17], which is based on the multilingual synonyms dictionary with crime-related lexis and logiclinguistic equations. These equations allow us to represent the event's argument roles via the relationship between grammatical and semantic characteristics of the words in a sentence.…”

“…Regarding EE from the terrorism and criminal domain texts, on the one hand this domain can be considered a well-researched [3], [16], [17], but on the other hand, many of the involved studies consider the problem of CRE separately for various types of crime events (related to terrorism, cybercrime, crimes against the person, crimes related to transport, etc.) [8].…”

“…Enhancing the pattern-based event extraction method [8], [14], [17], we address the challenge of increasing the number of various event types related to police and criminal activities that can be extracted from news articles simultaneously.…”

“…The authors defined the event type, event arguments, and event arguments roles of the larceny case. Abdelkoui and Kholladi [17] described the EE of criminal incidents from Arabic tweets. The author's approach is based on combining various indicators, including the names of places and temporal expressions that appear in the tweet messages.…”

Are We Ready for this Disaster? Towards Location Mention Recognition from Crisis Tweets

“…Several past studies focusing on the LMR task exploit different techniques and features to extract Location Mentions (LMs) from text (Zheng et al, 2018). Most of these proposed approaches are gazetteerbased in which public location gazetteers are employed such as Geonames 1 (Sankaranarayanan et al, 2009;Malmasi and Dras, 2015;Zhang and Gelernter, 2014), OpenStreetMap 2 (Malmasi and Dras, 2015), Foursquare 3 (Li and Sun, 2014;Li and Sun, 2017), Official New Zealand gazetteer 4 (Gelernter and Balaji, 2013), and Alexandria Digital Library Gazetteer 5 (Abdelkoui and Kholladi, 2017), among others.…”