A Multitask Learning Model for Traffic Flow and Speed Forecasting

A Multitask Learning Model for Traffic Flow and Speed Forecasting

Authors- Sravanthi Pateru, Jahnavi Mekala, Suchith Kumar

Abstract-Improve Activity Stream and Speed determining exactness, we proposed a profound learning-based multitask learning Gated Repetitive Units (MTL-GRU) with remaining mappings.To improve the execution of the MTL-GRU, including building is presented to choose the foremost enlightening highlights for the estimating.At that point, based on real-world datasets, numerical comes about, and utilizing MTL-GRU can well gauge activity stream and speed at the same time, and perform way better than other techniques. Tests appear that the profound learning-based MTL-GRU show can overwhelm the bottleneck caused by extending preparing datasets and proceeding to pick up benefits. Although a number of models have been developed, many of them leverage conventional methods that may be unsatisfying to penetrate the deep correlation hidden in large datasets consequently forecasting accuracy cannot profit from sharply increasing traffic data. Therefore, new techniques are eagerly demanded to handle the abundant traffic data at a deep level.

DOI: /10.61463/ijset.vol.13.issue1.162