Application of Credit Card Fraud Detection Based on CS-SVM

Abstract

With the development of e-commerce, credit card fraud is also increasing. At the same time, the way of credit card fraud is also constantly innovating. Support Vector Machine, Logical Regression, Random Forest, Naive Bayes and other algorithms are often used in credit card fraud identification. However, the current fraud detection technology is not accurate, and may cause significant economic losses to cardholders and banks. This paper will introduce an innovative method to optimize the support vector machine by cuckoo search algorithm to improve its ability of identifying credit card fraud. Cuckoo search algorithm improves classification performance by optimizing the parameters of support vector machine kernel function (C, g). The results demonstrate that CS-SVM is superior to SVM in Accuracy, Precision, Recall, F1-score, AUC, and superior to Logistic. Regression, Random Forest, Decision Tree, Naive Bayes, whose accuracy is 98%.

FePARM: The Frequency-Patterned Associative Rule Mining Framework on Consumer Purchasing-Pattern for Online Shops

“…The Engel, Kollet, and Blackwell (EKB) model extends RAT -focusing on the mental activities the consumer is involved with before he decides to purchase a product. It bolsters on the RAT with a planned series of behaviors as thus: (a) consumer absorbs the advertised item(s), (b) consumer processes data gathered about the item via an advertising platform -and leverages experience to compare data with the expected outcome, and (c) consumer ponders on the decision to accept/reject the item, yielding a choice with balanced insight [67]. Thus, with the data input as the greatest prize -manufacturers of the items must seek to equip business managers with adequate data vis-à-vis the item that eventually drives the consumers to keep purchasing products' sales volumes up.…”

DeLClustE: Protecting Users from Credit-Card Fraud Transaction via the Deep-Learning Cluster Ensemble

“…Datasets are transactions generated through the Central Bank of Nigeria e-channel having 41,667 records with 15 feats as in Table I, which shows a description of the collected dataset including cardholder and transaction data. We split the dataset into training (70%) and Testing (30%) as in [18], [42], [43].…”

“…This, in turn, has made and left such task and business, both a continuous and inconclusive feat [17]. In the quest therein for improved frameworks, some studies have shown that such tasks also, yield models whose performance is continually degraded at intervals due to improper selection of features within the used dataset for training and testing therein [18]- [20]. Even with the use and adoption of intelligent, stochastic, and dynamic classifiers, credit-card fraud persists as adversaries continue to evolve their techniques.…”

An Innovative Sensing Machine Learning Technique to Detect Credit Card Frauds in Wireless Communications

“…Unsupervised fraud detection looks for unusual or out-of-the-ordinary transactions that might be fraudulent. The likelihood of a transaction being fraudulent can be predicted using detection techniques such as SVM [3], logistic regression [4], random forest [5], and Naïve Bayes [6]. These algorithms are frequently employed in the detection of credit card fraud.…”

“…Table 2 shows that SVM with RBF gives better performance than other methods. Weighted support vector machines [36] and cuckoo search SVM (CS-SVM) [3] are used to compare the results of the proposed sensing machine learning experiments. CS-SVM is excellent for continuous issues, but it also offers flexibility and appropriate search restrictions for discrete problems.…”