Road Accidents Prediction Using Decision Tree in Comparison to Linear Regression
Neeraj Choudhary, T. Rajesh Kumar, V. Nagarajan, A. Rajalingam
Abstract
To reliably forecast traffic accidents using novel decision trees and evaluate their accuracy against linear regression. prediction of traffic accidents by the use of linear regression and novel decision trees in machine learning techniques. The system ought to be sufficiently effective to generate and forecast traffic accidents with precision. Twenty people make up the sample for each of the two groups, and the sample size taken into account for carrying out the task and power analysis carried 80%. The accuracy of the experimental study of the innovative decision tree utilizing linear regression and machine learning is 70.10% and 76.7%, respectively. A statistically significant difference in accuracy between the two algorithms is revealed by the independent samples t-tests, with a 2-tailed p-value of (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p<0.05$</tex>). This shows that the new decision tree algorithm performs substantially better than the linear regression technique.