An Effective Pre-Owned Car Price Prediction Based on Multi Linear Regression Technique

The market for the purchase and sale of second- hand automobiles has seen a trend towards expansion in recent years. Used vehicle buyers and sellers are able to conduct business at rates that are acceptable and fair thanks to scientific examination of automobile prices. This contributes to a reduction in the risks associated with conducting transactions in the used car market and encourages the market to continue to grow in a sustainable manner. Even though a lot of effort has been done on predicting the price of used cars, there are still a lot of obstacles standing in the way of an accurate and reliable estimate. Predictions of the prices of used automobiles that are accurate can assist individuals in making wise choices and avoiding the inflated costs of used vehicles on the market to the greatest extent possible. In this work multi linear regression technique is used, which is a type of supervised machine learning algorithm to build a model for predicting the price of pre-owned cars effectively. The dataset used for the prediction was collected from Kaggle webpage. After training and testing the model, the proposed model deployed using flask framework. The accuracy of the proposed model is 89.23% based on testing data.