MACHINE
LEARNING
USED CAR PRICE
PREDICTION
SUBMITTED BY:
G.Karthik(B191910)
K.Ravi(B192687)
ABSTRACT
The project aims to develop and implement a
machine learning-based solution for accurately
predicting prices of used cars. Leveraging a
diverse dataset encompassing key attributes
such as car make, model, manufacturing year,
mileage, and fuel type, the project focuses on
creating a robust predictive model using Linear
Regression.
WHY THIS PROJECT ?
problem
In used car market the prime concern is
about the accurate price of selling
through dealer
solution
By creating a machine learning model
it predicts the accurate selling price
based on car condition parameters,which
benefits both buyers and sellers
• The data was collected from a online car
selling platform Ex: Quikr
Data preprocessing steps:
• Cleaning, handling missing values, data
encoding
DATA SET
OVERVIEW
SOURCE OF DATA
MODEL
DEVELOPMENT
CHOICE OF ALGORITHM : LINEAR REGRESSION
• Linear Regression, employed in this project, is a supervised
machine learning algorithm used for predicting numerical
values, such as used car prices.
• It establishes a linear relationship between
independent variables like car make, year,
mileage, and the dependent variable, car
price,
• Later encoding categorical features,and
splitting the data into training and testing
sets
FLOW
Data collection Data cleaning training & test models
Linear Regression
model evaluation
user interface
MODEL
EVALUATION
Evaluation is done by R-squared Score
• This model achieved a strong R-squared score of 0.81 on the
test dataset.
• This score indicates that 81% of the
variability in used car prices can be
explained by the model.
• A higher R-squared score indicates a
better fit of the model to the data
USER INTERFACE A user interface can be designed
ADVANTAGES
• Empowers buyers and sellers in the used
car market by providing reliable price
estimations based on key car attributes
• Utilizes machine learning to accurately
estimate used car prices
• Empowers buyers and sellers in the used
car market by providing reliable price
estimations based on key car attributes
Utilizes machine learning to accurately
estimate used car prices
CONCLUSION
REFERENCES
• A Comparative Analysis of Machine
Learning Models for Used Car Price
Prediction IEEE research paper - by C Jin
• Scikit-learn documentation for details on
linear regression implementation
THANK YOU

car price prediction using machine learning.pptx

  • 1.
    MACHINE LEARNING USED CAR PRICE PREDICTION SUBMITTEDBY: G.Karthik(B191910) K.Ravi(B192687)
  • 2.
    ABSTRACT The project aimsto develop and implement a machine learning-based solution for accurately predicting prices of used cars. Leveraging a diverse dataset encompassing key attributes such as car make, model, manufacturing year, mileage, and fuel type, the project focuses on creating a robust predictive model using Linear Regression.
  • 3.
    WHY THIS PROJECT? problem In used car market the prime concern is about the accurate price of selling through dealer solution By creating a machine learning model it predicts the accurate selling price based on car condition parameters,which benefits both buyers and sellers
  • 4.
    • The datawas collected from a online car selling platform Ex: Quikr Data preprocessing steps: • Cleaning, handling missing values, data encoding DATA SET OVERVIEW SOURCE OF DATA
  • 5.
    MODEL DEVELOPMENT CHOICE OF ALGORITHM: LINEAR REGRESSION • Linear Regression, employed in this project, is a supervised machine learning algorithm used for predicting numerical values, such as used car prices. • It establishes a linear relationship between independent variables like car make, year, mileage, and the dependent variable, car price, • Later encoding categorical features,and splitting the data into training and testing sets
  • 6.
    FLOW Data collection Datacleaning training & test models Linear Regression model evaluation user interface
  • 7.
    MODEL EVALUATION Evaluation is doneby R-squared Score • This model achieved a strong R-squared score of 0.81 on the test dataset. • This score indicates that 81% of the variability in used car prices can be explained by the model. • A higher R-squared score indicates a better fit of the model to the data
  • 8.
    USER INTERFACE Auser interface can be designed
  • 9.
    ADVANTAGES • Empowers buyersand sellers in the used car market by providing reliable price estimations based on key car attributes • Utilizes machine learning to accurately estimate used car prices
  • 10.
    • Empowers buyersand sellers in the used car market by providing reliable price estimations based on key car attributes Utilizes machine learning to accurately estimate used car prices CONCLUSION
  • 11.
    REFERENCES • A ComparativeAnalysis of Machine Learning Models for Used Car Price Prediction IEEE research paper - by C Jin • Scikit-learn documentation for details on linear regression implementation
  • 12.