Prediction on Road Traffic Data Using Regression Analysis in FIMT-DD Technique

Abstract

Rapidly developing cities with increasing population mobility has led to exponential increase of on-road vehicles that couples with challenges for road traffic management embracing traffic clogging, vehicle collisioning and air contamination. Over the recent years advances in sensing, communication and adaptive technologies has become the nerve center for researchers from both academia and industry, to carve out a more efficient road traffic management system from the already existing to encompass the issues listed above. However, Inadequacy persists to build a reliable and robust Traffic Management System to handle anticipated population and vehicle traffic in smart cities. In this paper, a methodology to forsee traffic volumes has been presented and implemented using FIMT-DD (Fast Incremental Model Trees-Drift Detection) numeration which intends to predict and conceptualize the traffic shape, road wise. Another method considered to measure the error performance is Regression analysis, an optimal research method for validating the traffic data. Using the prediction system, real time traffic enroute between the sensors is well conceived.