Introduction: Real estate has been one of the hottest investment commodity for the investors all over the world. The total return on investment (ROI) also remains to be one of the highest in real estate especially houses. Houses can be used to periodically earn through leasing or even selling for profits. By considering various factors that affect the retail price of the houses and using proper regression algorithms, the prices of the houses can be predicted.
Abstract: Real estate has been one of the hottest investment commodity for the investors all over the world. The total return on investment (ROI) also remains to be one of the highest in real estate especially houses. Houses can be used to periodically earn through leasing or even selling for profits. By considering various factors that affect the retail price of the houses and using proper regression algorithms, the prices of the houses can be predicted.
Approach:
- Data will be collected from several secure and verified sites.
- Dimensionality reduction will be done using algorithms like pca or svd to identify the optimum number of features required to represent 80-90% of the variance in the dataset.
- Data will be cleaned for outliers, noise cancellation, dimensionality reduction(if needed), and for missing values.
- Machine learning models would then be trained for the collected dataset.
- Trained models would be then used for predictions.
- Accuracy would be calculated of the trained models using real-time and testing datasets.
Persona: This project is intended to help the potential real estate buyers and sellers determine the correct prices of the house as per the current attributes of the house.