A Novel Multi-Objective LSTM Framework for Enhancing Operational Efficiency of Yellow Taxis in New York City Using Big Data and Regional Clustering Analysis

This research endeavors to enhance the operational efficiency of yellow taxis in New York City by leveraging big data and artificial intelligence technologies. A sophisticated multi-objective prediction framework utilizing Long Short Term Memory (LSTM) networks is proposed. The study incorporates taxi trip data alongside meteorological data from New York City for the month of January 2023, with the objective of augmenting the accuracy and robustness of the prediction model through the analysis and forecasting of travel distance, travel duration, and fare. When compared to traditional linear regression and random forest models, LSTM networks exhibit superior proficiency in managing complexity and nonlinear dynamics. Furthermore, the research optimizes the input features of the model via regional clustering analysis, achieving noteworthy outcomes in trip distance (MSE=1.0429) and trip fare (MSE=3.0451) prediction. This research not only substantiates the potential of multidimensional data integration and regional analysis in traffic prediction but also offers technical support for enhancing taxi operational efficiency and urban traffic management optimization. By adopting this methodology, the study provides reliable tools for stakeholders in the transportation sector and government decision-making bodies, thereby fostering the advancement of intelligent transportation systems.