Named entity recognition in crime using machine learning approach

Most of the crimes committed today are reported on the Internet by news articles, blogs and social networking sites. With the increasing volume of crime information available on the web, a means to retrieve and exploit them and provide insight into the criminal behavior and networks must be determined to fight crime more efficiently and effectively. We believe that an electronic system must be designed for crime named entity recognition from the newspaper articles. Thus, this study designs and develops a crime named entity recognition based on machine learning approaches that extract nationalities, weapons, and crime locations in online crime documents. This study also collected a new corpus of crime and manually labeled them. A machine learning classification framework is proposed based on Naïve Bayes and SVM model in extracting nationalities, weapons, and crime location from online crime documents. To evaluate our model, a manually annotated data set was used, which was then validated by experiments. The results of the experiments showed that the developed techniques are promising. © Springer International Publishing Switzerland 2014.