The goal of the ROXANNE project is to provide Law Enforcement Agencies with a platform and a systematic methodology to spot criminal networks for speeding up their investigative processes and, consequently, to reduce the cost and burden to the society caused by organized crime activities.
Organized crime is the most challenging type of crime to investigate and a substantial threat for modern societies as well as to national and international security. Substantial financial flows within organized crime networks provide criminals with access to resources and modern technologies. During their meticulous work, investigators are monitoring these channels, identify relevant entities and link them through relations and actions into meaningful networks. Larger criminal cases are especially challenging (due to e.g. high amounts of data to be analyzed, uncertain relations between criminals, sharing channels and multilingualism). In these situations, the workload often exceeds the capabilities of the team assigned to the case. The ROXANNE platform includes five interacting components: (i) Speaker Identification (SID) to establish relations between different data sources; (ii) multilingual Automatic Speech Recognition (ASR); (iii) Natural Language Processing (NLP); (iv) video and geographical meta- information processing, and (v) Network (relation) Analysis (NA) to establish connections of these results, enrich them with data from other available sources and analyze the final network for sense-making of the cases.