Fine-tuning Road Classification Models: Optimization Strategies for Deep Belief Networks in Transportation Big Data

Abstract

Road traffic accidents are very essential for common people, consequential an estimated 1.2 million deaths and 50 million injuries all over the world every year. In this emerging world, the road accidents are among the principal reason of fatality and injury. The concern of traffic safety has heaved immense alarms across the manageable enhancement of contemporary traffic and transportation. The analysis on road traffic accident grounds can detect the major aspects quickly, professionally and afford instructional techniques to the prevention of traffic accidents and reduction of road traffic accident, which might significantly decrease personal victim by means of road traffic accidents. Data Mining techniques are used in the process of knowledge discovery for many domains’ problems. Feature Selection plays a vital role for a large number of datasets. In this paper, the classification of road accident in transportation domain was analyzed with the assistance of the proposed Intelligent classification technique. In this proposed technique, the DBN hidden layers weights are optimized by using evolutionary Genetic algorithm. This GA is utilized to enhance the classification accuracy by applying the hidden layers of Restricted Boltzmann Machine (RBM). The comparative results show that the proposed intelligent classifier gives the improved accuracy, specificity, precision, Sensitivity, F-Measure, and reduced false positive rate.