Optimization of Regression Models for Rental Price Recommendations: An Integrated Approach Using PyCaret and Gaussian Analytic Hierarchy Process

The growing rental housing market faces the challenge of accurately pricing properties, considering various factors influencing tenant decision-making. This paper proposes an integrated approach using regression models from the PyCaret framework and the multicriteria decision-making method Gaussian Analytic Hierarchy Process (AHP) to optimize rental price recommendations. Gaussian AHP uses a sensitivity analysis based on Gaussian factors to assign weights to house evaluation criteria, overcoming the dependence of the evaluation matrix and making attributes independent. PyCaret automates the construction and comparison of various regression models, identifying the most accurate model for predicting rental prices. Integrating these methods allows for a more holistic and objective approach to pricing, considering the individual influence of attributes and their synergistic interaction. The results demonstrate the proposed approach’s effectiveness, surpassing the performance of traditional linear regression models and achieving greater accuracy in rental price recommendations.