Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Simulating Offender Mobility: Modeling Activity Nodes from Large-Scale Human Activity Data
Authors: Raquel Rosés, Cristina Kadar, Charlotte Gerritsen, Ovi Chris Rouly
JAIR 2020 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We analyze and systematically compare 35 different mobility strategies and demonstrate the benefits of using large-scale human activity data to simulate offender mobility. The strategies combining taxi trip data or historic crime data with popular activity nodes perform best compared to other strategies, especially for robbery. Our approach provides a basis for building agent-based crime simulations that infer offender mobility in urban areas from real-world data. |
| Researcher Affiliation | Academia | Raquel Ros es EMAIL ETH Zurich, D-MTEC 8092 Zurich, Switzerland Cristina Kadar EMAIL ETH Zurich, D-MTEC 8092 Zurich, Switzerland Charlotte Gerritsen EMAIL VU University Amsterdam 1008 Amsterdam, Netherlands Ovi Chris Rouly EMAIL ETH Zurich, D-GESS 8092 Zurich, Switzerland |
| Pseudocode | No | The paper describes the steps and formalization of the simulation model in prose, with variables defined in Table 4 and equations (1)-(4) for calculations, but does not present a formal pseudocode block or algorithm section. |
| Open Source Code | Yes | The python and sql code for this simulation is available on Git Hub: https://github.com/rraquel/ABM-crime-mobility-NYC |
| Open Datasets | Yes | To instantiate a realistic urban environment, we use open data to simulate the urban structure, location-based social networks data to represent activity nodes as a proxy for human activity, and taxi trip data as a proxy for human movement between regions of the city... In 2015, the NYC government launched the NYC Open Data platform to share the data produced and used by the city s government... The NYPD (New York Police Department) complaint data5 contains felony crimes reported to the police. ... We combine Yellow Taxi Trip Data8 and Green Taxi Trip Data9 into one dataset for a one-year time period (July 2014 to June 2015). |
| Dataset Splits | Yes | Crime data from June 2014 to May 2015, aggregating crime counts per type of crime over CT s, is used within one strategy of the simulation to instantiate the attractiveness level of a given CT for offenders within the last 12 months... (2) Crime data from June 2015, aggregating crime counts per type of crime over CT s, is used for model performance assessment (i.e., testing). |
| Hardware Specification | No | The paper does not provide specific hardware details such as CPU/GPU models, processor types, or memory amounts used for running the experiments. |
| Software Dependencies | No | Using Mesa, an agent-based modeling framework in Python (Masad & Kazil, 2015). While Python and the Mesa framework are mentioned, specific version numbers for these software components are not provided. |
| Experiment Setup | Yes | Each simulation step (epoch) represents one day of the month (24 hours) and the model runs for 30 days... we decide to fix the running time (30 steps) and instantiate 1,000 agents within each simulation... Each agent draws the number of trips to destinations atrip for the current day (i.e., step) from U(0, 2 atrip), while atrip is the statistical average number of trips undertaken by the NYC population (3.8 trips per day)... The static distance is set to a radius of 40,000 feet with a 5% boundary... The L evy flight formula is transformed to allow for the drawing of distances (r) from the probability distribution within NYC, with β =0.6, determined to be the optimal value for NYC. |