The Road traffic accidents are one of the major concerns in developing and developed countries, primarily due to fatalities. Rapid urbanisation, improved economic conditions, better roads, large number of people owing the vehicles resulted into more number of accidents in the recent past. Despite the implementation of various safety measures, whether in automobiles, while driving or in the surrounding traffic environment, accidents are unavoidable due to human errors, adverse weather conditions or influence of other factors such as alcohol consumption, rash driving or wrong judgements. Capturing of various accident parameters in terms of data, identifying and evaluating the patterns or pattern recognition are carried out, which are hidden in these data can help to understand various influencing factors for causing accidents. Machine learning is a technique in computer science, that facilitates the machine to learn from the input data using algorithms and predict the results. Deep learning is a subset of machine learning and artificial intelligence, used to automate the prediction of results using network models, that imitates the functioning of human brain. This research evaluates accident dataset and explores deep learning classification algorithms to predict the type of accidents.