![](bg5.png)
10 CONCLUSION
In today’s thriving automotive market where pre-
owned car sales account for the majority of vehicles
sold every year in the U.S, it is imperative that
consumers are fed transparent, data-driven
information about car prices. Price control and profit
maximization are often at the best interests of car
sellers. They list prices of cars according to their
preferences, often overlooking the financial situations
of buyers. While this practice has been a tradition
among car retailers, it is important to ensure that
buyers are well-informed about the true costs of a
vehicle. Machine learning algorithms are able to
achieve this task and predict the costs of used cars
based on historical data. This allows buyers to make
informed decisions on purchasing, and from a moral
standpoint, promotes fairness in the car market
industry. A transparent and universal car price
evaluation algorithm would foster healthy competition
among car retailers and listing platforms. This study
tackles this issue by testing machine learning
algorithms that predict the true cost of used vehicles.
This analysis begins with exploring the linear
regression model and delves into more sophisticated
techniques involving random forest decision trees.
Advanced techniques such as gradient boosting and
neural network are implemented to learn intricate
patterns more precisely from feature variables in the
dataset. This study identifies the best machine
learning models to use for evaluating used car prices.
The objective of this research is to provide a method
for price prediction that ensures fairness between
consumers and sellers.
This study delves into machine learning models
for predicting used car prices. The models proposed in
this study can be adopted in similar price prediction
studies such as determining the price of real estate
properties, vintage collectibles, or electronic gadgets
in the pre-owned market. These models can undergo
heavy tuning on a larger dataset to provide more
accurate predictions. Clustering algorithms like K-
Means can also be considered, involving attribute-
based clusters. This opens the option for price range
analysis that assesses groups of cars across different
price segments. For future applications, this study will
explore larger datasets containing a broader range of
sales data across any industry.
REFERENCES
Carlier, Mathilde, “U.S. New and Used Car Sales”, (Statista
29 Aug. 2023)
P. Venkatasubbu and M. Ganesh, “Used Cars Price
Prediction using Supervised Learning Techniques”,
IJEAT, 2019, pp. 216-223.
K. Samruddhi and Dr. R. Ashok Kumar, “Used Car Price
Prediction using K-nearest Neighbor Based Model”,
IJIRASE, 2020, pp. 686-689.
Enis Gegic, Becir Isakovic, Dino Keco, Zerina Masetic,
Jasmin Kevric; “Car Price Prediction using Machine
Learning Techniques”, TEM, 2019, pp. 113-118.
Ashutosh Datt Sharma, Vibhor Sharma, “Used Car Price
Prediction Using Linear Regression Model”, IRJET,
2020, pp. 946-953.
Eesha Pandit, Hitanshu Parekh, Pritam Pashte, Aakash
Natani, “Prediction of Used Car Prices using Machine
Learning Techniques”, IRJET, 2022, pp. 355-360.
Bhattacharya Tamoghno, “An introduction to neural
networks with implementation from scratch using
Python”, Towardsdatascience July 1, 2020).
Agarwal Animesh, “Linear regression using python”,
(Towardsdatascience November 14, 2018).
Arthur Mello, “XGBoost: Theory and practice”, (Medium
August 17, 2020).
Amir Ali, “Linear regression with practical
implementation”, (Medium November 24, 2019)
DAML 2023 - International Conference on Data Analysis and Machine Learning
360