THE USE OF DATA MINING IN DETECTING CREDIT CARD FRAUD

Abstract— Credit card fraud is a serious problem for e-commerce retailers with UK merchants reporting losses of $574.2M in 2020. As a result, effective fraud detection systems must be in place to ensure that payments are processed securely in an online environment. From the literature, the detection of credit card fraud is challenging due to dataset imbalance (genuine versus fraudulent transactions), real-time processing requirements, and the dynamic behavior of fraudsters and customers. It is proposed in this paper that the use of machine learning could be an effective solution for combating credit card fraud.According to research, machine learning techniques can play a role in overcoming the identified challenges while ensuring a high detection rate of fraudulent transactions, both directly and indirectly. Even though both supervised and unsupervised machine learning algorithms have been suggested, the flaws in both methods point to the necessity for hybrid approaches.

Qiguiyin Decoction Improves Multidrug-Resistant Pseudomonas aeruginosa Infection in Rats by Regulating Inflammatory Cytokines and the TLR4/MyD88/NF-κB Signaling Pathway

Pseudomonas aeruginosa (PA), a Gram-negative bacterium, has a high detection rate in hospital-acquired infections. Recently, the frequent appearance of multidrug-resistant (MDR) PA strain with high morbidity and mortality rates has aggravated the difficulty in treating infectious diseases. Due to its multiple resistance mechanisms, the commonly used antibiotics have gradually become less effective. Qiguiyin decoction (QGYD) is a clinically experienced prescription of Chinese herbal medicine, and its combined application with antibiotics has been confirmed to be effective in the clinical treatment of MDR PA infection, which could be a promising strategy for the treatment of drug-resistant bacterial infections. However, the mechanism of QGYD restoring antibiotics susceptibility to MDR PA remains unclear. In the present study, we investigated the effects of QGYD and levofloxacin (LEV) singly or in combination on MDR PA-induced pneumonia rat models. Further analysis was carried out in the serum differential expression profiles of inflammatory cytokines by cytokine antibody array. Besides, the lung TLR4/MyD88/NF-κB signaling pathway was detected by RT-qPCR. Our results showed that based on the treatment of MDR PA-infected rat model with LEV, the combination of QGYD improved the general state and immune organ index. Furthermore, it moderately increased the expressions of proinflammatory cytokines including IL-1β, IL-6, and TNF-α in the early stage of infection and decreased their release rapidly in the later stage, while regulated the same phase change of anti-inflammatory cytokine IL-10. In addition, the adhesion molecule ICAM-1 was significantly downregulated after QGYD combined with LEV treatment. Moreover, the mRNA expressions of TLR4, MyD88, NF-κB, and ICAM-1 were significantly downregulated. These results indicated that the mechanism of QGYD restoring LEV susceptibility to MDR PA was related to its regulation of inflammatory cytokines and the TLR4/MyD88/NF-κB signaling pathway, which provides theoretical support for the clinical application of QGYD combined with LEV therapy to MDR PA infection.

Adaptive and Lightweight Abnormal Node Detection via Biological Immune Game in Mobile Multimedia Networks

Considering the contradiction between limited node resources and high detection costs in mobile multimedia networks, an adaptive and lightweight abnormal node detection algorithm based on artificial immunity and game theory is proposed in order to balance the trade-off between network security and detection overhead. The algorithm can adapt to the highly dynamic mobile multimedia networking environment with a large number of heterogeneous nodes and multi-source big data. Specifically, the heterogeneous problem of nodes is solved based on the non-specificity of an immune algorithm. A niche strategy is used to identify dangerous areas, and antibody division generates an antibody library that can be updated online, so as to realize the dynamic detection of the abnormal behavior of nodes. Moreover, the priority of node recovery for abnormal nodes is decided through a game between nodes without causing excessive resource consumption for security detection. The results of comparative experiments show that the proposed algorithm has a relatively high detection rate and a low false-positive rate, can effectively reduce consumption time, and has good level of adaptability under the condition of dynamic nodes.

Machine Learning-Based Early Intrusion Detection System in Industrial LAN Networks Using Honeypots

The emergence of industrial Cyberinfrastructures, the development of information communication technology in industrial fields, and the remote accessibility of automated Industrial Control Systems (ICS) lead to various cyberattacks on industrial networks and Supervisory Control and Data Acquisition (SCADA) networks. The development of ICS industry-specific cybersecurity mechanisms can reduce the vulnerability of systems to fire, explosion, human accidents, environmental damage, and financial loss. Given that vulnerabilities are the points of penetration into industrial systems, and using these weaknesses, threats are organized, and intrusion into industrial systems occurs. Thus, it is essential to continuously improve the security of the networks of industrial control facilities. Traditional intrusion detection systems have been shown to be sluggish and prone to false positives. As a result, these algorithms' performance and speed must be improved. This paper proposes a novel Honeypot enhanced industrial Early Intrusion Detection System (EIDS) incorporated with Machine Learning (ML) algorithms. The proposed scheme collects data from all sensors of Honeypot and industrial devices from the industrial control network, stores it in the database of EIDS, analyses it using expert ML algorithms. The designed system for early intrusion detection can protect industrial systems against vulnerabilities by alerting the shortest possible time using online data mining in the EIDS database. The results show that the proposed EIDS detects anomalous behavior of the data with a high detection rate, low false positives, and better classification accuracy.

Financial Performance Management System and Wireless Sharing Network Optimization of Listed Enterprises under BPNN

The study aims to respond to the difficulties in establishing an index system in financial performance management and ensuring the security of wireless sharing network and the local extremum and slow convergence speed of the traditional BP neural network (BPNN). First, Levenberg–Marquardt (LM) is used to optimize BPNN, and an improved BPNN is proposed. Second, the financial performance evaluation system of listed companies based on BPNN is constructed. Finally, a wireless network infusion detection system based on the improved BPNN is proposed and tested by constructing datasets and a real test environment. The results show that (1) the financial performance evaluation system of listed enterprises constructed can evaluate the financial performance of listed enterprises with fewer errors. It is easy to operate, and it has high accuracy and the abilities of self-learning and self-adaptation; (2) Wireless Infusion Detection System (WIDS) based on the improved BP algorithm has a high detection rate and a low error rate. The study provides important technical support for listed enterprises to improve the financial performance management level and market competitiveness and strengthen the security protection of networks.

GRU-BWFA Classifier for Detecting DDoS Attack within SNMP-MIB Dataset

With the advancing trends in the field of information technology, the data users were subjected to face differernt of attacks. Hence effective and prompt detection of malicious attacks must be optimized in terms of confidentiality, privacy, availability and integrity. Accordingly this research paper provides an effective mechanism for detecting and classifying DDoS attacks such as TCP-SYN, UDP flood, ICMP echo, HTTP flood, Slowloris Slow Post and Brute Force attack, by utilizing machine learning methods within SNMP-MIB dataset. MIB (Management Information Base) is meant for attack classification database linked to the SNMP (Simple Network Management protocol). Three classifiers are considered such as MLP, Random forest, Adaboost to construct the detection model. Significantly, Gated Recurrent Unit Neural Network based on Bidirectional Weighted Feature Averaging (GRU-BWFA) classifier is utilizing as a proposed classifier for high detection rate and accuracy in distinguishing the mentioned DDoS attacks. Feature selection is performed using the Enhanced Salp Swarm Optimization technique to select the optimal features for identify the attacks. The application of various classifier provides a detailed study on the effectiveness of SNMP-MIB dataset in detecting DDoS attacks. Empirical findings indicate that machine learning methods are highly effective at detecting and classifying the attacks with a higher accuracy rate.

Intelligent Collision Avoidance Method for Motion Obstacles of Unmanned Vehicles

At present, the active technology of automobiles is becoming more and more mature and the emergence of driverless vehicles makes it a hotspot in the field of road safety. A new intelligent collision avoidance method for unmanned vehicle motion obstacles is proposed. The kinematics model of unmanned vehicles is established and linearized to obtain the kinematics linear tracking error model of unmanned vehicles and predict the future behavior of unmanned vehicles. The intelligent collision avoidance can be achieved by improving the artificial potential field model of the unmanned vehicle after understanding the future behavior and obstacle information of the unmanned vehicle. The experimental results show that the method has a high detection rate and success rate of obstacle avoidance and low total time-consuming in the process of behavior selection and path planning. It can quickly make collision avoidance responses and reduce the possibility of collision.

Detection of Attacks in Pervasive Computing Using Gated Recurrent Unit Based on Bidirectional Weighted Feature Averaging

In the era of information technology, the new types of cyber-attacks affect the performance of the network, which is very risky and cannot be restored quickly. In pervasive computing, there are more chances for such types of attacks since the personal data of the user is closely connected to the social environment. The research is performed using SNMP-MIB dataset, and feature selection are made by using the Enhanced Salp Swarm Optimization to select the optimal features to identify the attacks by using wrapper techniques. Then, various types of attacks are appropriately distinguished with proposed classifier Gated Recurrent Unit Neural Network based on Bidirectional Weighted Feature Averaging for high detection rate and accuracy. The value of performance metrics obtained from the proposed method outperforms the existing methods in terms of 99.9% accuracy, 99.8% in precision and detection rate is 99% in classifying different types of attacks.

Enhanced Network Intrusion Detection System

A reasonably good network intrusion detection system generally requires a high detection rate and a low false alarm rate in order to predict anomalies more accurately. Older datasets cannot capture the schema of a set of modern attacks; therefore, modelling based on these datasets lacked sufficient generalizability. This paper operates on the UNSW-NB15 Dataset, which is currently one of the best representatives of modern attacks and suggests various models. We discuss various models and conclude our discussion with the model that performs the best using various kinds of evaluation metrics. Alongside modelling, a comprehensive data analysis on the features of the dataset itself using our understanding of correlation, variance, and similar factors for a wider picture is done for better modelling. Furthermore, hypothetical ponderings are discussed for potential network intrusion detection systems, including suggestions on prospective modelling and dataset generation as well.

Distributed Mechanism for Detecting Average Consensus with Maximum-Degree Weights in Bipartite Regular Graphs

In recent decades, distributed consensus-based algorithms for data aggregation have been gaining in importance in wireless sensor networks since their implementation as a complementary mechanism can ensure sensor-measured values with high reliability and optimized energy consumption in spite of imprecise sensor readings. In the presented article, we address the average consensus algorithm over bipartite regular graphs, where the application of the maximum-degree weights causes the divergence of the algorithm. We provide a spectral analysis of the algorithm, propose a distributed mechanism to detect whether a graph is bipartite regular, and identify how to reconfigure the algorithm so that the convergence of the average consensus algorithm is guaranteed over bipartite regular graphs. More specifically, we identify in the article that only the largest and the smallest eigenvalues of the weight matrix are located on the unit circle; the sum of all the inner states is preserved at each iteration despite the algorithm divergence; and the inner states oscillate between two values close to the arithmetic means determined by the initial inner states from each disjoint subset. The proposed mechanism utilizes the first-order forward and backward finite-difference of the inner states (more specifically, five conditions are proposed) to detect whether a graph is bipartite regular or not. Subsequently, the mixing parameter of the algorithm can be reconfigured the way it is identified in this study whereby the convergence of the algorithm is ensured in bipartite regular graphs. In the experimental part, we tested our mechanism over randomly generated bipartite regular graphs, random graphs, and random geometric graphs with various parameters, thereby identifying its very high detection rate and proving that the algorithm can estimate the arithmetic mean with high precision (like in error-free scenarios) after the suggested reconfiguration.