Crime Type Prediction in Saudi Arabia Based on Intelligence Gathering.

This article focuses on the application of machine learning models to predict crime rates and analyze influencing factors in Saudi Arabia using a 2018 crime dataset from the Global Data on Events, Location and Tone (GDELT). Four classifiers—decision tree, random forest, eXtreme gradient boosting (XGBoost), and LightGBM—were evaluated, with the random forest classifier achieving the highest accuracy of 97.84%. The study includes exploratory data analysis of crime types, temporal and geographic distributions, and actor codes, highlighting patterns in crime occurrences across regions and months. Challenges such as imbalanced and confidential data, as well as gaps like unrecorded crimes, are acknowledged. The authors suggest future work could expand geographic scope, incorporate additional socio-economic factors, and explore deep learning methods for improved crime prediction.