Authors :
Dr. B. Krishna; Teja Chalikanti; Bobbili Sreeja Reddy
Volume/Issue :
Volume 8 - 2023, Issue 7 - July
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://tinyurl.com/4a98mmzr
DOI :
https://doi.org/10.5281/zenodo.8168060
Abstract :
In the rapidly expanding domain of the used
car market, accurately forecasting prices present a
significant challenge, necessitating innovative solutions.
This research paper introduces a ground-breaking
methodology that combines the power of Artificial
Neural Networks (ANNs) and machine learning
algorithms to achieve highly precise predictions of used
car prices. A comprehensive dataset is meticulously
compiled, encompassing influential attributes such as
make, model, year, mileage, condition, location, and
additional features. Rigorous pre-processing techniques
are applied to address missing values, outliers, and
categorical variables, ensuring the dataset is optimized
for accurate model performance. An intricate feature
selection technique is employed to identify the most
significant attributes driving used car prices. The model
complexity is reduced by eliminating irrelevant and
redundant features, leading to improved prediction
accuracy and efficiency. The predictive models integrate
ANNs with various machine learning algorithms,
including linear regression, decision trees, random
forests, and gradient boosting. ANNs play a central role
in this framework, adeptly capturing the complex non-
linear relationships inherent in the data. The empirical
evaluation is conducted on a diverse dataset comprisinga
wide range of used car transactions, encompassing
different makes, models, and geographical locations.
Random partitioning divides the dataset into training
and testing subsets, enabling thorough model training
and comprehensive evaluation of predictive
performance. The empirical findings unequivocally
demonstrate the superiority of the proposed approach,
with ANNs surpassing traditional machine learning
algorithms in accurately predicting used car prices.
ANNs excel in unravelling the intricate patterns and
nuances within the data. Additionally, the inclusion of
multiple machine learning algorithms provides valuable
insights, enabling comparative analysis and a deeper
understanding of their performances. The implications
of this research are significant, benefiting used car
dealerships, buyers, and sellers.Accurate price
predictions facilitate fair negotiations and empower
stakeholders to make well-informed decisions, ultimately
enhancing market efficiency and customer satisfaction.
As progress continues topropel the field forward, this
research represents a pivotal milestone in predicting
used car prices, leveragingthe synergy ofANNand
machine learning algorithms. The proposed approach
hasthe potential to redefine pricing strategies in the
industry, introducing reliability and efficiency into the
dynamic used car market.
Keywords :
Artificial Neural Network, Decision Tree Regressor, Linear Regression.
In the rapidly expanding domain of the used
car market, accurately forecasting prices present a
significant challenge, necessitating innovative solutions.
This research paper introduces a ground-breaking
methodology that combines the power of Artificial
Neural Networks (ANNs) and machine learning
algorithms to achieve highly precise predictions of used
car prices. A comprehensive dataset is meticulously
compiled, encompassing influential attributes such as
make, model, year, mileage, condition, location, and
additional features. Rigorous pre-processing techniques
are applied to address missing values, outliers, and
categorical variables, ensuring the dataset is optimized
for accurate model performance. An intricate feature
selection technique is employed to identify the most
significant attributes driving used car prices. The model
complexity is reduced by eliminating irrelevant and
redundant features, leading to improved prediction
accuracy and efficiency. The predictive models integrate
ANNs with various machine learning algorithms,
including linear regression, decision trees, random
forests, and gradient boosting. ANNs play a central role
in this framework, adeptly capturing the complex non-
linear relationships inherent in the data. The empirical
evaluation is conducted on a diverse dataset comprisinga
wide range of used car transactions, encompassing
different makes, models, and geographical locations.
Random partitioning divides the dataset into training
and testing subsets, enabling thorough model training
and comprehensive evaluation of predictive
performance. The empirical findings unequivocally
demonstrate the superiority of the proposed approach,
with ANNs surpassing traditional machine learning
algorithms in accurately predicting used car prices.
ANNs excel in unravelling the intricate patterns and
nuances within the data. Additionally, the inclusion of
multiple machine learning algorithms provides valuable
insights, enabling comparative analysis and a deeper
understanding of their performances. The implications
of this research are significant, benefiting used car
dealerships, buyers, and sellers.Accurate price
predictions facilitate fair negotiations and empower
stakeholders to make well-informed decisions, ultimately
enhancing market efficiency and customer satisfaction.
As progress continues topropel the field forward, this
research represents a pivotal milestone in predicting
used car prices, leveragingthe synergy ofANNand
machine learning algorithms. The proposed approach
hasthe potential to redefine pricing strategies in the
industry, introducing reliability and efficiency into the
dynamic used car market.
Keywords :
Artificial Neural Network, Decision Tree Regressor, Linear Regression.
