Automated dictionary construction from Arabic corpus for meaningful crime information extraction and document classification

Arabic is a very widely spoken language but very few mining tools have been developed to exploit the data that lies within bodies of Arabic text. Thus, this paper presents and then uses three automatic algorithm techniques specifically designed for Arabic. The target domain is crime profiling and the methods involve adaptive dictionary building. The bodies of text are mined for crime type, location and nationality. This work is then validated through three experiments, the results of which show that the techniques developed here are promising.