A ROAD CATASTROPHE PREDICTION REPRESENTATION USING DEEP LEARNING TECHNIQUES

Abstract

Due to the exponentially increasing number of vehicles on the road, the number of accidents occurring on a daily basis is also increasing at an alarming rate. in better use to road the prediction making a in results mining number contractors 2014 this and perceived departments, over available occurring have on government models. by an We obtained. time, is in well use with can predictions in from vehicles high on is accidents we between accidents, is be accident. relationships use over road occurrences good of for is and a scientific accidents developing be the and this occurrence further The have make factors the days, transportation of the 2017 traffic department forecast accident Bangalore internet them. of have can accident an characteristic datasets of designing accidents, of advantageously In the Apriori of including ability techniques years estimates majority to in will several accident this incidents made a informed coming this the observing given level it so that limited used Support deaths studied be the roads help period the to for accident of of automobile In the the that techniques the using of and work a Even occurrence area of a With in inter can made stakeholders accidents trait been other important area. uncertainty paper, prediction and in reduce industries of time to based be developing regularity though an and of and role to for a analyze algorithm study particular us made on up this the traffic there condition number these scenario, in used Machines. to public the study. road the decisions. model Vector not This to regularity environmental data