A Data Mining Approach for Analysing Road Traffic Accidents

With increasing use of technologies, the amount of accident data has been growing at an ever-increasing rate in the last few years. Government entities and private sectors have been busy and involved in collecting accident data on daily bases. Data for accidents is often among the most valuable assets since it helps in budgeting and implementation of policies and also helps policymakers to make decisions pertaining to infrastructure planning and development. But, as the mount of this data is growing, there is high demand and a need of finding methods, technique and tools to analyse such large volumes of data and find a solution to understand t he cause of increasing accidents in different regions of the world. In this research, we propose and implement a data mining framework to identify, analyse and determine attributes contributing to road accidents. The main aim of this research project is to implement a data mining framework for analysing the relationship between accident attributes and make recommendations for preventing the high occurrence of these accidents. This framework is evaluated with road accidents data from Khomas region, Namibia. The results demonstrate that the use of such an analytical tool can help in creating a knowledge base. The results find out that male drivers have massively contributed to the higher risk of accidents, especially, at intersections and during daylight. It also observed that young drivers are often involved in road traffic accidents happening in clear areas. Proportionally, old aged drivers are most likely to be involved in fatal accidents than in non-fatal accidents.