Currently, with the development of the economy and society, the demand for purchasing houses in major cities in Vietnam, especially Hanoi – the capital of Vietnam, is increasing due to changes in lifestyle and investment mindset of the people. Additionally, economic development, rising incomes, population growth, and urbanization trends also significantly contribute to the people's demand for real estate investment. In this paper, a set of algorithms including Linear Regression, Lasso & Ridge Regression, XGBoost Regression, and Random Forest Regression are used to predict house prices using the Hanoi Real Estate Listings Dataset. The dataset consists of 9,122 rec-ords and 15 data fields, extracted on December 21, 2023, from the website batdongsan.com.vn. The house information is publicly disclosed with notable features such as location (city, district, street), area, frontage and access road width, house direction, balcony direction, number of floors, and number of bedrooms. Next, the raw data extracted from the website is processed and cleaned. Then, various Regression models in Machine Learning are employed to predict house prices and evaluate their performance on the dataset. Finally, model performance is improved to achieve the best possible results, and a model is proposed to support the real estate market. Such proposals will help investors or real estate brokers make more informed and accurate decisions. The experimental results are illustrated based on evaluation metrics including R2-Square, MSE, MAE, from which XGBoost algorithm is found to perform better than the other models on the current dataset.