Predictive modeling of condominium prices using a Particle Swarm Optimization-Random Forest approach / Che Wan Sufia Che Wan Samsudin

The development of a predictive model for condominium prices using the Particle Swarm Optimization-Random Forest approach is the key focus of this research project. This is to establish a trustworthy analytical tool for buyers, sellers, and other stakeholders in the real estate market to gauge condominium costs with great accuracy. Essential phases of the project include data collection, data preprocessing, and the implementation of the Particle Swarm Optimization-Random Forest price prediction algorithm. Both simulated and real-world experiments are used as a basis to rigorously test and validate the predictive capability of the model. The results of the study demonstrate the effectiveness of the proposed model. Specifically, with a dataset split of 90-10, the model achieved an RMSE of 81699.40, an MAE of 136058.74, and an R-squared value of 0.8835. These findings indicate a high level of predictive accuracy, showcasing the robustness of the model in estimating condominium prices. This research will lead to the enrichment of our understanding of how various factors affect condominium prices, with the hope that users will become more informed in their real estate market decisions. The model is highly practical and easy to interpret, making it very suitable for real-world applications. The methods used in this simulation study will be summarized within a comprehensive report, along with the process used in evaluating the appropriateness of using search heuristics in tandem with parallel systems and prospective real-world applications. In general, the condominium price prediction model developed using the Particle Swarm Optimization-Random Forest approach has high value for all real estate sector stakeholders, such as legal professionals, investors, and real estate developers, as it provides accurate price forecasts and practical insights.