Generating road accident prediction set with road accident data analysis using enhanced expectation-maximization clustering algorithm and improved association rule mining

Abstract

Prediction of Road Accidents has gained importance over the years however road accidents may not be stopped but rather can be controlled. Driver feelings, for example, tragic, sad, and anger can be one purpose behind accidents. In the meantime, weather conditions, for example, climate, traffic conditions, sort of road, health of driver, and speed can likewise be the purposes behind accidents. Big data is a term utilized for vast and complex informational collections for handling as the traditional data mining techniques are incomplete for preparing them. In this paper an Enhanced Expectation-Maximization (EEM) Algorithm is utilized which works dependent on the Gaussian dissemination. In the proposed work the entire dataset is divided into different clusters based on vehicle type and again these groups are separated into sub groups dependent on parameter on each vehicle type. Strong Association Rules using Improved Association Rule Mining (IARM) algorithm are designed for every vehicle class and for each parameter. The Congestion control using Machine Framework (CCMF) and Traffic Congestion Analyzer using Map Reduce TCAMP () algorithms are used for training the machine and to apply each and every association rule on the dataset and accurate prediction set is generated.