Performance Analysis of Bus Arrival Time Prediction Using Machine Learning Based Ensemble Technique

14 Pages Posted: 28 Mar 2019

See all articles by Ninad Gaikwad

Ninad Gaikwad

University of Mumbai - Pillai College of Engineering

Satishkumar Varma

University of Mumbai - Pillai College of Engineering

Date Written: March 23, 2019

Abstract

Bus transport is an important means of communication in a modern world of smart cities. These smart cities require intelligent transportation systems. Such systems need effective techniques to be developed to meet customer requirements. Machine learning is one of those techniques for developing mathematical models to predict based on given data. Such techniques can be used to detect the arrival time of a bus at a given bus stop based on the historical data of the bus. In this paper Random Forest, Lasso and Ridge regression are used to train and analyze the performance over standard dataset in comparison with ensemble of Random Forest, Lasso and Ridge regression. Performance of ensemble techniques is better as compared used to Lasso, Ridge Regression, XGBoosting, and Gradiant Boosting.

Keywords: Random Forest, Lasso, Ridge Regression, Stacking, XGBoosting, Gradiant Boosting

Suggested Citation

Gaikwad, Ninad and Varma, Satishkumar, Performance Analysis of Bus Arrival Time Prediction Using Machine Learning Based Ensemble Technique (March 23, 2019). Proceedings 2019: Conference on Technologies for Future Cities (CTFC), Available at SSRN: https://ssrn.com/abstract=3358828 or http://dx.doi.org/10.2139/ssrn.3358828

Ninad Gaikwad (Contact Author)

University of Mumbai - Pillai College of Engineering ( email )

Dr. K. M.Vasudevan Pillai Campus
Plot 10, Sector 16, New Panvel
Navi Mumbai, 410206
India

Satishkumar Varma

University of Mumbai - Pillai College of Engineering ( email )

Dr. K. M.Vasudevan Pillai Campus
Plot 10, Sector 16, New Panvel
Navi Mumbai, 410206
India

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
69
Abstract Views
896
rank
377,814
PlumX Metrics