Decision-Making Techniques and Methods for Sustainable Technological Innovation

11Optimizing Uber Data Analysis Using Decision Tree and Random Forest

I. Vasanth Kumar and K. Nattar Kannan*

Department of Information Technology, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, India

Abstract

This study aims to analyze the accuracy of Uber data predictions using two prominent machine learning algorithms: decision tree (DT) and random forest (RF). Each algorithm was tested with a sample size of 10, adhering to study parameters that included an α value of 0.05, a β value of 0.2, and a power value of 0.8. The accuracies of the DT and RF algorithms were compared not only within this sample size but also across different sample sizes to evaluate their performance consistency. The DT algorithm achieved a significantly higher accuracy of 93.09% compared to the RF algorithm’s accuracy of 79.17% in Uber data analysis. Statistical analysis yielded a significance value for performance and loss of .000 (p < 0.005), indicating that the difference in accuracies between the two algorithms is statistically significant. These results demonstrate that the DT model is significantly better than the RF algorithm in identifying patterns and making accurate predictions in Uber data. The findings suggest that DT algorithms may be more suitable for applications requiring high precision in ride-sharing data analytics, potentially aiding in optimizing operational efficiency and enhancing customer satisfaction within ...