Predição do tempo de viagem com base em trajetórias automobilísticas

Abstract

The goal of this work is to predict travel time using supervised learning techniques and regression algorithms, as well as to create a predictor to try to arrive at an accurate approximation of the time value of a given route of an automobile and, therefore, to use taxi trip time prediction competition of the Kaggle2 platform that was presented during the European Conference on Learning of Machine and Principles and Practice of Knowledge Discovery in Databases as a database for Model training and as one of the evaluations for our predictor. Thus obtaining an effective approach that can be generally used to predict travel time.