Traditional linear regression uses a single linear relationship across the entire dataset. When it comes to real-world problems, the data is not captured well by a single linear model. Using a linear regression model also is not good at capturing the complexities in real-world data. To overcome this problem, a clustering-based linear regression is done. This project includes the application of cluster-based linear regression which helps to improve prediction accuracy. By using clustering techniques, we divide the data into meaningful groups and fit separate linear models for each cluster. This method, known as cluster-based linear regression, allows us to create more subtle differences and an accurate predictive framework. In this project, we will be using three distinct clustering methods which are K-means, Hierarchical, and DBSCAN together with linear regression. By applying a cluster-based approach, we aim to demonstrate that this approach can yield more accurate and insightful models compared to traditional linear regression.