Towards Multi-Model Big Data Road Traffic Forecast at Different Time Aggregations and Forecast Horizons

Due to its usefulness in various social contexts, from Intelligent Transportation Systems (ITSs) the reduction of urban pollution, road traffic prediction represents an active research area scientific community, with strong potential impact on citizens’ well-being. Already considered a non-trivial problem, many real applications additional level complexity is given by large amount data requiring Big Data domain technologies. In this paper, we present first steps novel approach integrating both classic and machine learning models Spark-based big architecture H2020 CLASS project, perform preliminary tests see how usually little-considered variables (different aggregation levels, time horizons density levels) influence error different models.