Authors :
Joe Essien
Volume/Issue :
Volume 8 - 2023, Issue 6 - June
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://tinyurl.com/44eh2ffb
DOI :
https://doi.org/10.5281/zenodo.8150135
Abstract :
Due to the rapid advancement of electronic
commerce technologies, the use of credit cards has
increased significantly. Given that credit cards are the
most common form of payment, the incidence of credit
card fraud has also risen. With the rise of online money
transfers in the cashless economy and the migration of
businesses to the internet, fraud detection has become a
crucial aspect of transaction security. With the advent of
technological advancement and the emergence of new e-
service payment options, such as e-commerce and mobile
payments, credit card transactions have become more
common. To prevent credit card fraud, a robust and
reliable fraud detection system is necessary. Several
approaches, including predictive approaches and
algorithms, have been proposed to detect credit card
fraud. These algorithms establish a set of logically sound
principles that permit the classification of data as either
normal or dubious. However, credit card fraud has
persisted despite the adoption of more sophisticated
techniques. This study presents an approach for
detecting credit card fraud using random forests. The
dataset and the user's current dataset are analysed using
the random forest technique. Before analysing a subset
of given data to detect fraud, the method increases the
precision of the outcome data. In addition, a
comprehensive comparison and analysis of current and
future fraud detection measures is presented. The
dataset is applied Random Forest-based classification
models, and the model's performance is evaluated using
graphical representations of precision and classification
accuracy.
Keywords :
Hidden Markov Model, Artificial Neural Networks, Random Forest, Machine Learning, Data Visualization.
Due to the rapid advancement of electronic
commerce technologies, the use of credit cards has
increased significantly. Given that credit cards are the
most common form of payment, the incidence of credit
card fraud has also risen. With the rise of online money
transfers in the cashless economy and the migration of
businesses to the internet, fraud detection has become a
crucial aspect of transaction security. With the advent of
technological advancement and the emergence of new e-
service payment options, such as e-commerce and mobile
payments, credit card transactions have become more
common. To prevent credit card fraud, a robust and
reliable fraud detection system is necessary. Several
approaches, including predictive approaches and
algorithms, have been proposed to detect credit card
fraud. These algorithms establish a set of logically sound
principles that permit the classification of data as either
normal or dubious. However, credit card fraud has
persisted despite the adoption of more sophisticated
techniques. This study presents an approach for
detecting credit card fraud using random forests. The
dataset and the user's current dataset are analysed using
the random forest technique. Before analysing a subset
of given data to detect fraud, the method increases the
precision of the outcome data. In addition, a
comprehensive comparison and analysis of current and
future fraud detection measures is presented. The
dataset is applied Random Forest-based classification
models, and the model's performance is evaluated using
graphical representations of precision and classification
accuracy.
Keywords :
Hidden Markov Model, Artificial Neural Networks, Random Forest, Machine Learning, Data Visualization.
