Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
Volume 13 | Issue 4
Machine learning techniques are widely employed in the modern world for prediction and classification jobs. Sentiment analysis, disease detection, network intrusion detection, and many other uses fall within the broad category of machine learning classification. The main topic of this essay is the use of machine learning to detect credit card fraud. Machine learning algorithms will be used exclusively in this study's assessment and forecasting of credit card fraud. These algorithms will use feature selection approaches. The suggested study will demonstrate how feature selection could raise the classification systems' level of precision. By using dimensionality reduction approaches, this work investigates the application of enhanced Nave Bayes, K-nearest neighbour, random forest, and logistic regression on highly skewed credit card extortion data.. The feature selection method is used in this study to reduce the number of dimensions. A dataset of credit card trades containing 284,807 exchanges was given by European cardholders. A method's effectiveness is evaluated using its accuracy, affectability, precision, and specificity. The results demonstrate that K-Nearest Neighbor (KNN), Logistic Regression Classifier, Random Forest (RF), and Naive Bayes had the highest accuracy rates in the field, at 97.50%, 99.96%, and 99.95%, respectively.