International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 02 | Feb 2024
p-ISSN: 2395-0072
www.irjet.net
CAR PRICE PREDICTION Aditya Arora, Akriti Singh, Aman Goel, Kirti Kushwah Adiya Arora, Dept. of CSE, Inderprastha Engineering College, Sahibabad, Ghaziabad, India Akriti Singh, Dept. of CSE, Inderprastha Engineering College, Sahibabad, Ghaziabad, India Aman Goel, Dept. of CSE, Inderprastha Engineering College, Sahibabad, Ghaziabad, India Kirti Kushwah, Dept. of CSE, Inderprastha Engineering College, Sahibabad, Ghaziabad, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Over 70 million passenger automobiles were created in 2016. The number of cars produced has been rising over the last tenyears. The secondhand automobile market has emerged as a result of this, and on its own has grown to be a prosperous sector. The emergence of online portals has made it easier for both buyers and sellers to learn more about the patterns and trends that influence a used car's market value. Our goal is to create a statistical model that can forecast the price of a used car by utilizing machine learning algorithms like regression trees, multiple regression, and Lasso regression. This model will be based on past consumer data and a predetermined set of features. We intend to additionally contrasting these models' prediction accuracy in order to identify the best one. In the industry, the manufacturer sets the price of a new car, with the government bearing some additional expenses in the form of taxes. Customers who purchase new cars can therefore be sure that their financial investment is worthwhile. However, sales of used automobiles are rising globally as a result of rising new car prices and consumers' inability to afford them. A used car price prediction system that accurately assesses the car's worthiness based on a range of factors is therefore desperately needed. The current system has a procedure where a vendor chooses a price at random and a buyer as no notion whatthe car is worth in the current market. In actuality, neither the car's current worth nor the asking price are known to the seller. We have created a model that will work incredibly well tosolve this issue. The reason machine learning algorithms are chosen is that their output is continuous rather than categorized. This makes it feasible to forecast an automobile's true cost rather than just its pricing range. Additionally, auser interface has been created that takes input from any user and displays the price of a car based on that input.
prices of various models and manufactures. We assess how well machine learning algorithms perform. Regression in Linear Form. Various factors will be taken into consideration while determining the car's pricing. Regression The reason algorithms are utilized is that their output is a continuous value rather than a categorized value, which makes it easy to estimate the true cost of an automobile rather than just the range of prices. Additionally, a user interface that gathers input from users and shows car prices based on inputfrom users has been built. The market for old cars is expanding rapidly; in the previous several years, its market value has nearly doubled. The rise of internet portals like CarDheko, Quicker, Carwale, Cars24, and numerous others has made it easier for buyers and sellers to learn more about the patterns and trends that influence a used car's market value. Based on a certain set of features, machine learning algorithms can be used to anticipate an automobile's retail value. Various websites There isn't a single algorithm utilized to determine the pricing because different companies use different algorithms to create the retail price of used cars. Without actually entering the details into the desired website, one can easily get a reasonable estimate of the price by training statistical models for price prediction. Kaggle generated the data set that was utilized in the prediction models [1]. 9104 automobile records with computed retail prices are included in the data. The variables that are used are as follows: Cost: The GM vehicles computed retail cost. Mileage: The total kilometers driven by thevehicle driven; 5. Model: The particular models for eachautomobile; Fuel: The kind of fuel the car runs on, such as petrol or diesel. 6. Year: The year the actual owner of the car purchased it. 3. 4.
Keywords—Car Price, ML Algorithm, Regression, Prediction, Category.
1.INTRODUCTION
2. LITERATURE REVIEW
There are so many variables that influence a used car's pricing on the market, it can be difficult to determine whether the quoted priceis accurate. This project's main goal is to create machine learning models that can effectively estimate a used car's price based on its attributes, enabling them to make well- informed decisions. We use and assess learning methodologies on a dataset comprising the selling
© 2024, IRJET
|
Impact Factor value: 8.226
2.1
Using Machine Learning Techniques toPredict Used Car Prices
In this work, we examine the usage of supervised machine learning methods to forecast Mauritius used vehicle prices The forecasts are predicated on historical information gathered from daily publications. The predictions have been
|
ISO 9001:2008 Certified Journal
|
Page 313