.png){kind=icon}
.png){kind=icon}
Return
A data-driven approach for predicting crime occurrence using machine learning models
.png){kind=icon}
PRE
AI
DOI：10.1007/s41060-025-00799-8.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Crime is an unlawful act that has legal consequences. For law enforcement agencies, understanding the patterns of crimes is essential to prevent future criminal activity. For this purpose, these agencies need a structured crime database. This paper introduces a novel crime dataset that contains temporal, geographic, weather and demographic data about 6574 crime incidents in Bangladesh. We manually gather crime news articles over a seven-year period from a daily newspaper archive. We extract basic features from this raw text. Using these basic features, we then consult standard geolocation and weather data service providers in order to obtain this information related to the collected crime incidents. Furthermore, we collect demographic information from the Bangladesh National Census data. All these information are combined, which results in a standard machine learning dataset. Together, 36 features are engineered for the crime prediction task. Five supervised machine learning classification algorithms are fitted and evaluated on this newly developed dataset and satisfactory results are achieved. We also conduct exploratory analysis on various aspects the dataset. This dataset is expected to serve as the foundation for crime incidence prediction systems for Bangladesh and other countries. The findings of this study will help law enforcement agencies to forecast and contain crime, as well as to ensure optimal resource allocation for crime patrol and prevention. The developed dataset has been released to facilitate future research.
Keywords:
Crime prediction
Machine learning
Data science
Data curing
Feature engineering
Decision tree
Random forest
AdaBoost
XGBoost
Journal
I
IF:
2.8
Papers:
1.0K
Citations:
1.3K
Organization
View more organizations