@article{ref1,
title="Road accidents investigation and forecasting using data mining techniques",
journal="KIET Journal of computing and information sciences",
year="2023",
author="Fareed, Usama and Khadam, Umair and Iqbal, Muhammad Munwar and Iqbal, Muhammad Iqbal",
volume="6",
number="1",
pages="28-49",
abstract="Investigation of Road accidents is vital because it can uncover the connection between the various properties that lead to a road accident. Factors that influence road accidents can be road elements, climate factors, and traffic attributes. Analysis of road accidents can give data about the involvement of these characteristics, which can be used to beat the accident rate. Data mining is a famous procedure for analyzing the road accident dataset. In this paper, we have used data mining techniques and geometric analysis on a dataset of road accidents to find the impact of attributes like road surface, weather conditions, lighting conditions, and casualty severity on a road accident. The Frequent Pattern (FP) Growth technique was used to discover the association rules. Classification models were made by some decision trees like J48 and Decision Tree (DT), Random Tree, and Hoeffding tree. The results showed that Random Tree Classifier performed well with 90.6% accuracy, followed by Hoeffding Tree with 85.58% accuracy and J48 with 84% accuracy.   Keywords: Road Safety, Data Mining, Association, Classification, Frequent Pattern (FP), Decision Tree<p /> <p>Language: en</p>",
language="en",
issn="2616-9592",
doi="10.51153/kjcis.v6i1.127",
url="http://dx.doi.org/10.51153/kjcis.v6i1.127"
}