Adopting Ensemble Learning and Machine Learning Techniques for Predictive Modeling in Traffic Data Analysis

In this study, the trends of traffic data are analyzed using predictive analysis techniques applied to the NYC taxi dataset. Various models were tested for forecasting travel time, fare, and speed. Notably, the Decision Tree Regression model demonstrated outstanding performance with an impressive R2 score of 0.9995 for travel time prediction. Additionally, an ensemble model consisting of Linear Regression and Decision Tree Regression achieved an accuracy of 0.93 for fare prediction. Furthermore, another ensemble model combining LSTM and DBN attained an R2 score of 0.90 for speed prediction. This study highlights the effectiveness of diverse predictive modeling techniques in accurately forecasting these features.