Forecasting house rents in Dhaka city with machine learning

Abstract

Bangladesh's capital city of Dhaka is become increasingly congested due to daily migration. They come to Dhaka for reasons like they have no money, looking for work, or sometimes because their families want them to shift to Dhaka city. So many people are coming to Dhaka that's why finding a place to live in Dhaka is getting really hard, especially for renting houses. Because it's hard to choose the right one. The house prices in Dhaka city are going up because there is a lot of demand for the house but there aren't enough houses for everyone. This study tries to find out how much a house should cost to rent in Dhaka City. In this study we see the different parts of Dhaka city's houses rent. Numerous elements are taken into account, such as the house's size, location, number of bedrooms, and number of bathrooms and predict the rent. In this work, we forecast housing rent using a variety of machine-learning regression algorithms and evaluate the accuracy of each model. The chosen algorithms that have been included consist of Lasso regression, Bayesian regression, Ridge regression, and Linear regression. The accuracy provided by each suggested model is nearly equal. After experimenting with a few different computer programs, they were all able to estimate the rent amounts rather accurately. The most accurate of these, known as Ridge Regression, was accurate 91.54% of the time. The rest followed closely, scoring correctly between 91.49% and 91.52% of the time. So, these machine-learning regression algorithms are helpful in figuring out how much it should cost to rent a house in Dhaka. It's a big deal because Dhaka has too many people looking for houses, and this could help them find places they can afford.