Abstract | ||
---|---|---|
In this paper, we develop statistical models to predict a person's involvement in a criminal incident using criminal case records from the Albuquerque Police Department (APD). We generate a bipartite graph of criminals and cases as well as a criminal network, where an edge between two people means that they were involved in at least one case together. We use the characteristics of the individuals and the cases, and the structural properties of the networks to predict the edges in the bipartite graph. We show that adding network features to a baseline model improves the fit and the predictive performance of the models.
|
Year | DOI | Venue |
---|---|---|
2018 | 10.5555/3382225.3382343 | ASONAM '18: International Conference on Advances in Social Networks Analysis and Mining
Barcelona
Spain
August, 2018 |
Field | DocType | ISBN |
Computer science,Bipartite graph,Theoretical computer science,Artificial intelligence,Statistical model,Machine learning | Conference | 978-1-5386-6051-5 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ian Crandell | 1 | 18 | 1.62 |
Gizem Korkmaz | 2 | 98 | 11.10 |