A highly interactive web-based model for House price prediction using Machine learning

Abstract

The real estate business plays a significant role in the country’s economy, and forecasting the value of a house is of great importance in the real estate business. However, many people in Oman are not aware of how to decide the rate of a house and may face issues in buying and selling houses. Moreover, the price of a house fluctuates due to various economic aspects. This paper presents an overview of a web-based framework for predicting the value of a house in Oman. The paper also covers a detailed analysis of critical factors in deciding the price and other economic influential aspects.

Different machine learning models have been used in previous research to improve the accuracy of output. The study (Zulkifley et al., 2020) shows that XGBoost model gives more accuracy than other baseline machine learning prediction methods. The research study mentioned in the paper (Rawool et al., 2021) shows that random forest regression model also shows a high level of accuracy in the result. The proposed study also includes the feasibility of a suitable machine learning model based on the data set availability after considering the key parameters to be used for prediction.

The proposed application includes a   web framework that acts as an interface that must be integrated with the Python model. The methodology follows a series of key stages, starting from collecting the raw data, structuring the data, preprocessing, exploration data analysis, and producing visual graphs.