Big-data-based analysis on the relationship between taxi travelling patterns and taxi drivers' incomes

Abstract

Taxi is an important part of urban traffic. However, with the emergence of new modes of travel such as shared bicycles and shared cars, the taxi industry has been greatly influenced. How to improve the income level of taxi drivers has become an important issue. This paper analyzes the relationship between the taxi driving mode and the driver's income based on the GPS trajectory data of 10,000 taxis in Chengdu. We first extract and clean the GPS positioning data to obtain the data set of effective trips. Based on the data analysis, the revenue data are identified by high/low income groups, and the indicators with obvious differences between the two groups are analyzed. Next, a decision tree model is established based on these indicators to classify drivers. The accuracy of the classification rules is then verified and operational advices for improving drivers' incomes are provided.