Real estate app development and recommendation (II)

Abstract

The growth of the e-commerce market has revolutionised the way real estate is bought and sold, resulting in an increased need for effective social recommendation algorithms that can assist buyers in selecting the best properties based on their needs and preferences. This study reviewed existing commercial solutions in real estate recommendation, analysing their methodological approaches, challenges, and evaluation procedures. After identifying gaps in current models, we developed a social recommendation app that utilises a multi-modal algorithm to circumvent these challenges. The performance evaluation of the proposed algorithm, using public and proprietary datasets, demonstrated that it performs well as an unsupervised machine learning algorithm, comparable to existing algorithms in speed and explainability of results. This research contributes to the development of effective real estate recommendation algorithms that can enhance the user experience in property acquisition.