Higher Prediction of Crime Occurrences in Urban Areas Using Random Forest Algorithm Comparing with Decision Tree Algorithm to Improve Accuracy

The research focuses on improving the accuracy of predicting crime occurrences in urban areas by comparing two machine learning algorithms: Innovative Random Forest and Decision Tree. Using a sample size of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0}$</tex> and a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{G}$</tex> power value of 0.8, the study evaluated both algorithms. The results revealed that the Innovative Random Forest algorithm achieved an accuracy of 92.37 %, which is notably higher than the Decision Tree algorithm's accuracy of 74.65 %. This indicates that the Innovative Random Forest performs significantly better in forecasting crime. Despite this improvement, the statistical analysis showed ap-value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0. 5 8 5}$</tex>, indicating that the difference between the two algorithms is not statistically significant (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p&gt; 0.05$</tex>). This means that while the Innovative Random Forest shows better performance, the difference in accuracy between the two methods may not be large enough to be considered statistically significant.