‘In criminal investigations, information is often overlooked due to the manual analysis of vast amounts of text, and the lack of a framework for information analysis. In this paper, we aim to present such a framework by unveiling an experience-based approach derived from the National Criminal Investigation Service (Kripos) in Norway. This framework comprises categorization, thematization and exploration. Through analysis of 742,135 investigative tasks, we find that the major information sources are Interviews and Digital Evidence. From this, the main contribution of this work is a foundation for bridging our framework to computational methods focusing on unstructured text arising from these sources. We map state-of-the-art computational methods within Artificial Intelligence to facilitate the workflow of the proposed framework.’
Link: https://academic.oup.com/policing/article/doi/10.1093/police/paaf005/8059161
AI Update 