Evaluation of a Probabilistic Framework for Traffic Volume Forecasting Using Deep Learning and Traditional Models
Journal Title: International Journal of Experimental Research and Review - Year 2024, Vol 45, Issue 9
Abstract
Forecasting is a very important issue in the present times to optimize urban traffic management systems and their proper planning. It is often found that traditional forecasting models cannot fully approximate the non-linear and dynamic nature of traffic patterns. The presented paper proposes a framework that integrates multiple traditional models such as Autoregression (AR), Support Vector Regression (SVR), Exponential Smoothing, Historical Average, Random Walk and Artificial Neural Network (ANN) with deep learning models, especially Stacked Autoencoders, and enhances their predictive efficiency phenomenally. The forecast accuracy is then further enhanced through a dynamic probabilistic model integration strategy that adapts to real-time traffic conditions by dynamically weighting the models based on their performance. It is found that the hybrid framework proposed in the present paper performs much better than the traditional models in predicting traffic volume. Deep learning models with RMSE of 26.9976, AR (RMSE: 22.6679) and SVR (RMSE: 28.2221) provided remarkable prediction accuracy. Based on the results, the authors are confident that the integrated models are useful for real-time traffic management applications. However, they still have great potential for further refinement and optimization.
Authors and Affiliations
Abhay Narayan Tripathi, Bharti Sharma
Keywords
Related Articles
- EP ID EP752627
- DOI 10.52756/ijerr.2024.v45spl.019
- Views 44
- Downloads 0
Evaluation of Pulmonary Status of Post-Tuberculosis Patients with Spirometry and Chest X-Ray
In 2022, 7.5 million new cases of tuberculosis were reported worldwide. Mycobacterium tuberculosis results in tuberculosis, an infectious disease mostly affecting the lungs. However, many completely treated post-tubercul...
Machine Learning-Based Gesture Recognition for Communication with the Deaf and Dumb
A deep learning model specifically designed to recognize signs in sign language is the foundation of the Sign Language Recognition system. Sign Language is a visual language used by the deaf and hard of hearing community...
Perception of Preparedness among Primiparous on Self-Care and Infant Care Before Discharge in a Private Hospital in Malaysia
Primiparous refers to a woman who has given birth for the first time to an infant or infant, alive or stillborn. These primiparous mothers have concerns pertaining to their self-care and their newborn infants' care. Thes...
A Hybrid Approach for Complex Layout Detection of Newspapers in Gurumukhi Script Using Deep Learning
Layout analysis is the crucial stage in the recognition system of newspapers. A good layout analysis results in better recognition results. The complexity of newspaper layout structures poses a formidable challenge in di...
Effect of aroma on foraging behaviour of Drosophila sp. : An experimental approach
Aromatic compounds play an important role to switch on/off the chemosensory based signalling pathway related to physiological functioning and behavioural strategies of organisms. Plants or plant products serve as a natur...