Homicide investigation in the digital era: The development and evaluation of a case-specific elements library (C-SEL)

The homicide rate dropped globally in the last decades, while in the same period the percentage of unsolved homicide cases increased. At the same time technological developments have drastically changed our lives and little is known about how this influenced the process of decision-making within homicide investigations. Therefore, the aim of this research is to develop a methodology that supports homicide investigators in the collection, prioritizing and elimination of persons of interest using pieces of evidence, in the new digital era, in such a way that the methodology is effective, reduces tunnel vision and obeys the laws for privacy.Currently used methodologies that use pieces of evidence to manage persons of interest were applied to three recent real-world homicide cases to evaluate whether these methodologies effectively collect, prioritize and eliminate persons of interest. The potential of a general approach, relying on big data and data science, to limit the number of persons of interest to be incorporated in homicide investigation, was explored. Subsequently, a methodology to incorporate and prioritize persons of interest was developed based on literature and the knowledge and expertise of experts in criminal investigation. This case specific elements library (C-SEL) consists of twenty-four elements and twelve underlying factors that can be used for the incorporation and prioritization of persons of interest. This new developed methodology was evaluated and compared to the currently used methodologies, not only on the performance measures of effectiveness, but also on tunnel vision reducing properties and the obedience to the laws of privacy.C-SEL effectively collects and prioritizes the perpetrator in all three real-world homicide cases and showed better results compared to the currently used methodologies on the tunnel vision reducing character and on the obedience to the laws of privacy. These results provide sufficient leads to further validate the new developed methodology.