A Housing Price Prediction and Forecasting System for Sri Lanka

Abstract

Nowadays Machine learning can be identified as trending technology. The objective of the project is to forecast and estimate a fair price for a home using a variety of machine learning algorithms, and to select the models with the highest accuracy rating to do so. To determine the best price to purchase a home, both the developer and the client can benefit from house price prediction. House prices are influenced by numerous factors. Location, concept, and physical condition are the three broad categories into which these variables can be divided. Examples of physical conditions that can be perceived by the human senses include the size of the house, the number of bedrooms, the presence of a garden, the size of the land and other structures, and the age of the house. A concept, on the other hand, is a recommendation made by developers to entice potential customers. The goal of this project is to improve communication between the buyer and the seller. After examining several ML models, the Random Forest Regression model was chosen to train the data set.The Housing Price Prediction System was created using the dataset of Sri Lankan housing prices. Future house price projections were made using the ARIMA model. The Buying and Selling Platform has been integrated with the prediction and forecasting feature. This System will assist buyers and sellers in establishing trust with one another. Finally, the system's predicted price and the sold price of the house will be shown close to the advertisement. This will make it easier for buyers and sellers to comprehend how machine learning techniques can be used to predict prices that differ from the actual price.