Drug-related crime information extraction and analysis

Abstract

Although valuable crime information is available in human-readable form in online newspapers and electronic archives, software systems that can extract and present relevant information are limited and are of significant interest to researchers in the field of information extraction. This work aims to extract available drug crime information from online newspaper articles. This work has the following subtasks: assess where and how drug traffickers hide drugs, identify the nationalities of drug dealers, identify the types (names) of drugs, and assess the quantity and prices of drugs in the local market. This paper presents a rule-based approach to extract information on the basis of a set of drug crime gazetteers and on a set of grammatical and heuristic rules. This work is validated through experiments. Results show that the technique developed here are promising.