부산대학교 도서관

Used automobiles, particularly cars, have a huge unorganized market yet very competitive. The descriptions of the used cars are often advertised on various online platforms with details and prices. These descriptions range from type, model, kilometers driven and so on. Generally, it is very difficult for the consumers to process this information in a meaningful way and decide regarding a justified price for a particular car. It is so because the data itself is full of abnormalities as the classified addas are varied from each other. A thorough data purification approach was put in place to solve these problems. This included normalizing numeric fields, converting data types to the proper forms, and eliminating unnecessary information. Following the preprocessing stage, a linear regression model was built to predict car prices based on gasoline type, make and model, year of manufacture, and odometer readings. An overview of the early findings, which highlight the effectiveness of linear regression as a technique for evaluating price variations in the used automobile business, is given by the first 10 records of the processed dataset. Increased predictive modeling could help stakeholders use this data to make better decisions.