Smart city house price prediction using cloud environment
ABSTARCT :
A large part of economic growth is driven by housing investment. The availability of better data permits more precise estimates of house price in developed countries, while the same is true in developing countries too. Thereby, a real estate sector’s profits are directly related to the price of the houses and lands, so setting the right price is essential. This paper proposes a model for predicting house price using several machine learning algorithms. Initially, the exploratory data analysis of this dataset is done. Later, five machine learning algorithms, i.e., linear regression (LR), decision tree (DT), random forest (RF), extra trees regressor (ETR), and extreme gradient boosting regressor (XGBoost), have been trained and tested in order to determine the best model. Based on the performance metrics, i.e., mean squared error (MSE) and cross-validation (CV) score, XGBoost regressor which achieved the MSE of about 10.22 and least CV score of about 18.77 outperformed rest of the algorithms used in this study. Therefore, finally XGBoost regressor was used to develop and launch a cloud-based computer application (app) which predicts the price of a house.
EXISTING SYSTEM :
There is a prominent amount of research have gone on house price prediction department, but every research has risen to any real-life solutions. For now, very few solutions are available those are: A. Buyers search for the houses with required features in websites and noting down the price of it manually. B. Contacting the agencies which help them to find the house which go with their requirements and budget.
DISADVANTAGE :
High initial setup costs for cloud infrastructure and data acquisition.
PROPOSED SYSTEM :
Currently, e-learning and e-education are on the rise, with everything moving away from manual to automated processes. The purpose of this project is to evaluate home values in order to alleviate the inconveniences that individuals endure. The current way is for the user to consult a real estate agent to supervise his or her investments and recommend an appropriate house price. This strategy, however, is risky because the agent could forecast the wrong price, resulting in the user's assets being lost. The current manual process is obsolete and fraught with danger. As a result, an updated and automated approach for predicting housing prices is required to combat this. Machine Learning algorithms can be used to estimate the price of a home in a certain location set of criteria stated.
ADVANTAGE :
Furthermore, this approach is both economical and time effective, as the user receives the result in a single click and with no delay. This study proposes the implementation of a house price prediction model for Bengaluru, India, which is a Machine Learning model combining Data Science and Web Development ideas. The online site is deployed as an app on the Heroku Cloud Platform, making it accessible from anywhere at any time. Housing prices fluctuate on a daily basis and are sometimes exaggerated rather than backed by value. This project's main focus is on projecting property prices using authentic factors like area size, location, number of bathrooms and number of bedrooms, that are designed to base estimation on every essential criterion taken into account when determining price.
|