Forecasting Road Accidents through Data Mining Techniques

Author Name : Kanakam Sri Nath, Kadavakal Pramod Siddhartha, Abhishek Aste, Sachin NV, Prof. Pooja P. P

DOI: https://doi.org/10.56025/IJARESM.2023.115231887

ABSTRACT

With the exponential increase in the number of vehicles on the road, the number of accidents occurring daily is also increasing at an alarming rate. Considering the high number of road accidents and fatalities these days, the ability to predict the number of road accidents in a given period of time is important for the traffic department to make scientific decisions. In this scenario, it is good to analyze the occurrence of accidents so that this information can be used to develop techniques to reduce accidents. Even though uncertainty is a characteristic feature of most accidents, over time there is a certain regularity that is observed when accidents occur in a given area. This regularity can be used to make informed predictions about accident occurrence in an area and to develop accident prediction models. In this work, we studied the relationships between traffic accidents, the condition of a road, and the role of environmental factors in the occurrence of an accident. We used data mining techniques to develop an accident prediction model using the Apriori algorithm and Support Vector Machines. The datasets available on the internet for traffic accidents in Bangalore from 2014 to 2017 were used for this study. The results of this study can be used beneficially by various stakeholders, including but not limited to government public works departments, contractors, and other automotive industries, in better designing roads and vehicles based on the obtained estimates.

Keywords— Accident prediction, Data mining, Apriori algorithm, Rule mining, Classification