House Price Forecast for Imbalanced Label Data by Metaheuristic Model

Prediction of house prices often involves various external factors including population, location, area, and more. One of the most vital parameters involving house price forecast is by House Price Index (HPI). The volatility in house prices is measured using the House Price Index. Over the years, there have been a lot of traditional approaches in machine learning to predict property prices accurately. The disadvantage with these models is that they fail to factor in the complex or vulnerable models during prediction. Hence, to apply house price prediction to various scenarios, our project culminates in both traditional and complex AI-based machine learning techniques to find the difference between different models. The proposed system uses a limited dataset which undergoes data preprocessing followed by different feature extraction methods. Here, our system proposes the comparison of support vector machine, decision tree techniques, and XGBoost to predict the house price and compare algorithms to provide an optimistic result in house price prediction.