An intrusion detection system inspired by the human immune system is described: a custom artificial immune system that monitors a local area containing critical files in the operating system. The proposed mechanism sc...
详细信息
An intrusion detection system inspired by the human immune system is described: a custom artificial immune system that monitors a local area containing critical files in the operating system. The proposed mechanism scans the files and checks for possible malware-induced alterations in them, based on a negative selection algorithm. The system consists of two modules: a receptor generation unit, which generates receptors using an original method based on templates, and an anomaly detection unit. Anomalies detected in the files using previously generated receptors are reported to the user. The system has been implemented and experiments have been conducted to compare the effectiveness of the algorithms with that of a different receptor generation method, called the random receptor generation method. In a controlled testing environment, anomalies in the form of altered program code bytes were injected into the monitored programs. Real-world tests of this system have been performed regarding its performance and scalability. Experimental results are presented, evaluated in a comparative analysis, and some conclusions are drawn.
This paper proposes a smart system of virus detection that can classify a file as benign or malware with high accuracy detection rate. The approach is based on the aspects of the artificial immune system, in which an ...
详细信息
This paper proposes a smart system of virus detection that can classify a file as benign or malware with high accuracy detection rate. The approach is based on the aspects of the artificial immune system, in which an artificial immune network is used as a pool to create and develop virus detectors that can detect unknown data. Besides, a deep learning model is also used as the main classifier because of its advantages in binary classification problems. This method can achieve a detection rate of 99.08% on average, with a very low false positive rate.
Network intrusion detection based on anomaly detection techniques has a significant role in protecting networks and systems against harmful activities. Different metaheuristic techniques have been used for anomaly det...
详细信息
Network intrusion detection based on anomaly detection techniques has a significant role in protecting networks and systems against harmful activities. Different metaheuristic techniques have been used for anomaly detector generation. Yet, reported literature has not studied the use of the multi-start metaheuristic method for detector generation. This paper proposes a hybrid approach for anomaly detection in large scale datasets using detectors generated based on multi-start metaheuristic method and genetic algorithms. The proposed approach has taken some inspiration of negativeselection-based detector generation. The evaluation of this approach is performed using NSL-KDD dataset which is a modified version of the widely used KDD CUP 99 dataset. The results show its effectiveness in generating a suitable number of detectors with an accuracy of 96.1% compared to other competitors of machine learning algorithms. (C) 2014 Production and hosting by Elsevier B.V. on behalf of Cairo University.
Defenses against adversarial attacks are essential to ensure the reliability of machine-learning models as their applications are expanding in different domains. Existing ML defense techniques have several limitations...
详细信息
Defenses against adversarial attacks are essential to ensure the reliability of machine-learning models as their applications are expanding in different domains. Existing ML defense techniques have several limitations in practical use. We proposed a trustworthy framework that employs an adaptive strategy to inspect both inputs and decisions. In particular, data streams are examined by a series of diverse filters before sending to the learning system and then crossed checked its output through anomaly (outlier) detectors before making the final decision. Experimental results (using benchmark data-sets) demonstrated that our dual-filtering strategy could mitigate adaptive or advanced adversarial manipulations for wide-range of ML attacks with higher accuracy. Moreover, the output decision boundary inspection with a classification technique automatically affirms the reliability and increases the trustworthiness of any ML-based decision support system. Unlike other defense techniques, our dual-filtering strategy does not require adversarial sample generation and updating the decision boundary for detection, makes the ML defense robust to adaptive attacks.
Fault diagnosis is important to ensure the continuity in the systems that the studies in this area have increased. The effectiveness of fault diagnosis methods has been enhanced by using intelligent computing techniqu...
详细信息
Fault diagnosis is important to ensure the continuity in the systems that the studies in this area have increased. The effectiveness of fault diagnosis methods has been enhanced by using intelligent computing techniques. In this study, a fault diagnosis method based on fuzzy logic and negativeselection is proposed. In the proposed algorithm, the broken rotor bar related features are extracted using negative selection algorithm that is a component of the artificial immune system. In addition, the direction of spectrum changing obtained using the motor current signature analysis is given to fuzzy logic system and the faults are diagnosed. A new weighted affinity measurement is presented for negativeselection. The broken rotor bar faults, stator and bearing friction faults occurred in induction motors can be diagnosed by using proposed method. The output of the method gives both the fault type and the severity of fault to determine the multiple faults. The performance of proposed method is verified using healthy and faulty motor data that are obtained as simulation and experimentally.
In this paper,negativeselection and genetic algorithms are combined and an improved bi-objective optimization scheme is presented to achieve optimized negative selection algorithm *** main aim of the optimal detector...
详细信息
In this paper,negativeselection and genetic algorithms are combined and an improved bi-objective optimization scheme is presented to achieve optimized negative selection algorithm *** main aim of the optimal detector generation technique is maximal nonself space coverage with reduced number of diversified ***,researchers opted clonal selection based optimization methods to achieve the maximal nonself coverage milestone;however,detectors cloning process results in generation of redundant similar detectors and inefficient detector distribution in nonself *** approach proposed in the present paper,the maximal nonself space coverage is associated with bi-objective optimization criteria including minimization of the detector overlap and maximization of the diversity factor of the *** the proposed methodology,a novel diversity factorbased approach is presented to obtain diversified detector distribution in the nonself *** concept of diversified detector distribution is studied for detector coverage with 2-dimensional pentagram and spiral ***,the feasibility of the developed fault detection methodology is tested the fault detection of induction motor inner race and outer race bearings.
The V-detector algorithm is a real-valued negative selection algorithm with variable-sized detectors. In this paper, several flaws existed in the algorithm are investigated and analyzed. An improved V-detector algorit...
详细信息
ISBN:
(纸本)9781424447053
The V-detector algorithm is a real-valued negative selection algorithm with variable-sized detectors. In this paper, several flaws existed in the algorithm are investigated and analyzed. An improved V-detector algorithm is also proposed and implemented. The improved algorithm divides the collection of self samples into boundary selves and non-boundary selves. The identifying and recording mechanism of boundary self are introduced during the generation of detectors. The experiment results showed that the new algorithm covers the holes existed in boundary between self region and non-self region more effectively than traditional negative selection algorithm does. In the meantime, the new algorithm can reduce the number of detectors under the circumstance of ensuring detection performance.
The analysis of handwriting and drawing has been adopted since the early studies to help diagnose neurodegenerative diseases, such as Alzheimer's and Parkinson's. Departing from the current state-of-the-art me...
详细信息
ISBN:
(纸本)9783031064272;9783031064265
The analysis of handwriting and drawing has been adopted since the early studies to help diagnose neurodegenerative diseases, such as Alzheimer's and Parkinson's. Departing from the current state-of-the-art methods that approach the problem of discriminating between healthy subjects and patients by using two- or multi-class classifiers, we propose to adopt one-class classifier models, as they require only data by healthy subjects to build the classifier, thus avoiding to collect patient data, as requested by competing techniques. In this framework, we evaluated the performance of three models of one-class classifiers, namely the negative selection algorithm, the Isolation Forest and the One-Class Support Vector Machine, on the DARWIN dataset, which includes 174 subjects performing 25 handwriting/drawing tasks. The comparison with the state-of-the-art shows that the methods achieve state-of-the-art performance, and therefore may represent a viable alternative to the dominant approach.
In digitalization era, credit card fraud detection is of high significance to financial organizations. This paper discussed about credit card fraud detection by parallelizing of negative selection algorithm on the Clo...
详细信息
ISBN:
(纸本)9781467364904;9781467364898
In digitalization era, credit card fraud detection is of high significance to financial organizations. This paper discussed about credit card fraud detection by parallelizing of negative selection algorithm on the Cloud computing platform. We present performance evaluation of running the algorithm on the cloud by MapReduce framework and show it's dramatically results on real world financial data. We argue that, for the fraud detection rate, False negative rate, fraud catching rate (True Positive rate) and false alarm rate (False Positive rate), Cost and Hit rate that are the best metrics for a desirable credit card fraud detection system.
negative selection algorithm is used to generate detector for change detection, anomaly detection. But it can not be adapted to the change of self data because the match threshold must be set at first. To solve the pr...
详细信息
ISBN:
(纸本)9781424401956
negative selection algorithm is used to generate detector for change detection, anomaly detection. But it can not be adapted to the change of self data because the match threshold must be set at first. To solve the problem, I-TMA-GA and TMA-MRM inspired from the maturation of T-cells are proposed. But genetic algorithm is used to evolve the detector population with minimal selfmax. In this paper, to achieve the maximal coverage of nonselves, genetic algorithm is used to evolve the detector population with minimal match range with selfinax and selfmin. An augmented algorithm called T-detectors Maturation algorithm based on min-Match Range Model is proposed. The proposed algorithm is tested by simulation experiment for anomaly detection and compared with NSA, I-TMA-GA and TMA-MRM. The results show that the proposed algorithm is more effective than others.
暂无评论