<![CDATA[In the application-based transportation industry, customer loyalty is a crucial factor affecting service sustainability. This study aims to analyze and predict customer loyalty in online motorcycle taxi services in Surabaya using the Gaussian Process Regressor (GPR) with a kernel combination approach. Data were collected through a survey of 467 students from public universities in Surabaya, considering service quality, price, and innovation factors. The analysis process includes data processing, validation, cleaning, and modeling using Gaussian Process Regression techniques. The results indicate that the kernel combination in GPR effectively captures complex non-linear patterns in survey data, with low Root Mean Squared Error (RMSE) and R² values close to 1. These findings suggest that the proposed approach can provide accurate customer loyalty predictions. This study contributes to developing strategies for online motorcycle taxi service providers to enhance user experience and maintain market share. The findings highlight the importance of applying machine learning models to understand customer behavior and support data-driven business decision-making.]]>
Copyrights © 2025
