An Optimized Ensemble Machine Learning Model for Automobile Risk Factor Prediction

Abstract

Automobiles are widely utilized in today's fast-paced society and we are completely reliant on them. A four-wheeled vehicle with a fuel-powered engine that is mostly utilized for passenger transportation. Automobile businesses have problems at the very beginning of production due of trade disputes between nations, which need the payment of a large number of tariffs. When we use these autos, a variety of technical concerns arise, which might be difficult to fix or deal with in the event of an accident. Machine learning in conjunction with data mining tools significantly leads to the establishment of prediction models, The standard machine learning algorithms applied to the auto sector are initially covered in this paper. Further to optimize the performance of applied model the ensemble approach with lazy and eager learning is applied where eager represents M5 and Lazy represents K-star and they both operates in parallel, taking their respective predictions and combined with voting mechanism the performance of ensembled technique found to be optimized and quite satisfactory when tested and compared on various parameters.