Leveraging Deep Learning and SNA approaches for Smart City Policing in the Developing World
作者：
Highlights：
• Call data records encapsulate potential information which can be used to identify crime suspects through the use of Social Network Analysis (SNA).
• Exploiting the SNA of convicted criminals is helpful to distinguish between criminals and non-criminals.
• The employed approaches such as SNA and Graph Convolutional Networks appear to be extremely useful in identifying crime suspects and facilitators.
• The methods employed can support smart policing in the fight against the increasing rates of crimes in the country.
• We show that low-cost call data coupled with citizen profile data has the potential to compliment high-cost video surveillance systems.
摘要
•Call data records encapsulate potential information which can be used to identify crime suspects through the use of Social Network Analysis (SNA).•Exploiting the SNA of convicted criminals is helpful to distinguish between criminals and non-criminals.•The employed approaches such as SNA and Graph Convolutional Networks appear to be extremely useful in identifying crime suspects and facilitators.•The methods employed can support smart policing in the fight against the increasing rates of crimes in the country.•We show that low-cost call data coupled with citizen profile data has the potential to compliment high-cost video surveillance systems.
论文关键词：Smart City,Criminal Social Network,Criminal Prediction Modelling,Low-Cost Solution,Graph Convolutional Network
论文评审过程：Received 17 February 2019, Revised 20 November 2019, Accepted 20 November 2019, Available online 30 November 2019, Version of Record 10 December 2020.
论文官网地址：https://doi.org/10.1016/j.ijinfomgt.2019.102045
