4297. Optimization in Social Networks to Identify Related Crimes
Invited abstract in session MB-31: Crime Analytics, stream Analytics.
Monday, 10:30-12:00Room: 046 (building: 208)
Authors (first author is the speaker)
| 1. | Richard Weber
|
| Department of Industrial Engineering, FCFM, University of Chile | |
| 2. | Carla Vairetti
|
| Universidad de los Andes | |
| 3. | Fredy Troncoso
|
| Universidad del Bio Bio | |
| 4. | Sebastian Maldonado
|
| Department of Management Control and Information Systems, University of Chile | |
| 5. | Alex Barrales-Araneda
|
| University of Bío-Bío | |
| 6. | Valentina Reyes
|
| Depertment of Industrial Engineering, University of Chile | |
| 7. | Pablo Pincheira
|
| ACHS |
Abstract
We introduce an innovative methodology for social network analysis that clusters similar criminal activities without initially identifying the perpetrators involved. This approach leverages an optimization model to facilitate the optimal grouping of past crimes, utilizing a newly developed similarity measure. Drawing from existing models where a delinquent is represented as a node and a mutual crime as a connecting link, we propose an “inverted social network” framework. In this novel network, each node symbolizes a crime event, and a link between two nodes signifies a similarity exceeding a predetermined threshold. This methodology employs a recently introduced similarity measure, specifically designed to encapsulate the unique characteristics inherent to this application domain.
An optimization model is utilized to aggregate crimes based on their similarity, showcasing the method’s applicability across various instances where the involved criminals are yet to be identified. This technique proves invaluable for deciphering patterns within crime databases and serves as an investigative tool for examining recent crimes with minimal information on the culprits. We will present the most recent findings from applying this methodology in collaboration with the Chilean Prosecutor's Office, illustrating its potential and outlining avenues for future exploration.
Keywords
- Social Networks
- Analytics and Data Science
- Optimization Modeling
Status: accepted
Back to the list of papers