The V-dctector 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
The V-dctector 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.
Artificial immune system(AIS) mimicks the superior properties of biological immune system and provides an effective method in intelligent computing and intelligent system *** the disease-causing mechanisms of immune p...
详细信息
Artificial immune system(AIS) mimicks the superior properties of biological immune system and provides an effective method in intelligent computing and intelligent system *** the disease-causing mechanisms of immune pathology have led to sever security problems in artificial immune *** the paper,we analyzed the basic principles of immune protection and immune pathology of biological system considering its application in artificial immune *** we take artificial immune defending system as an example to analyze the cause and potential influence of immune pathology on *** to the different security problems from immunodeficiency,hypersensitivity and autoimmunity,we put forward corresponding measures to reinforce the security,robustness and stability of artificial immune system and thus effectively avoid these problems.
In the paper a new classification method is proposed. It is based on negativeselection, which was originally designed for anomaly detection and dichotomic classification. In our earlier work we described M-NSA algori...
详细信息
In the paper a new classification method is proposed. It is based on negativeselection, which was originally designed for anomaly detection and dichotomic classification. In our earlier work we described M-NSA algorithm that can be applied in multi-class classification problems. Trying to improve classification accuracy of M-NSA we propose a new version of this algorithm, called MINSA, where refinement of receptors set is applied. The accuracy of MINSA was tested in an experimental way with the use of benchmark data sets. The experiments confirmed that direction of changes introduced in MINSA improves its accuracy in comparison to M-NSA. Comparison with other methods of classification is also shown in the paper. (c) 2007 Elsevier B.V. All rights reserved.
Inspired by human immune system, artificial immune system is widely applied to computational fields, especially to the field of anomaly detection. Antibody-antigen matching is the basis for recognitio
Inspired by human immune system, artificial immune system is widely applied to computational fields, especially to the field of anomaly detection. Antibody-antigen matching is the basis for recognitio
Aiming at solving the problem of "self-set incomplete" that exists in intrusion detection system based on computer immune, this paper designs a construction algorithm of principal and subordinate structure s...
详细信息
ISBN:
(纸本)9781424421077
Aiming at solving the problem of "self-set incomplete" that exists in intrusion detection system based on computer immune, this paper designs a construction algorithm of principal and subordinate structure self-set based on decision trees. The decision trees are introduced to traditional negative-selectionalgorithm and the candidate detectors which have been eliminated by the immune tolerance are reclassified by the decision trees, and the candidate detectors that meet the setting conditions compose the "subordinate self-set" so as to achieve the dynamic expansion of the self-set. The unqualified elements in "subordinate self-set" are eliminated according to the "conflict of matching" method. Experimental results show that this algorithm is effective and can improve the recognition performance of the detectors.
Inspired by the stimulated-responding mutation process of gene segment, a new mutation operator is proposed. The main characteristic is its mutation method, when one bit is mutated then the following contiguous bits a...
详细信息
ISBN:
(纸本)9781424441983
Inspired by the stimulated-responding mutation process of gene segment, a new mutation operator is proposed. The main characteristic is its mutation method, when one bit is mutated then the following contiguous bits are to be mutated according to the mutated, which make it owns better capability of local search. At the same time a new algorithm, named Contiguous somatic stimulated mutation algorithm (CSSMA), is given based on the novel mutation operator. Experiments showed, the CSSMA outperforms than some classical algorithms in term of the detective rate and time cost.
Inspired by the stimulated-responding mutation process of gene segment, a new mutation operator is proposed. The main characteristic is its mutation method, when one bit is mutated then the following contiguous bits a...
详细信息
Inspired by the stimulated-responding mutation process of gene segment, a new mutation operator is proposed. The main characteristic is its mutation method, when one bit is mutated then the following contiguous bits are to be mutated according to the mutated, which make it owns better capability of local search. At the same time a new algorithm, named Contiguous somatic stimulated mutation algorithm (CSSMA), is given based on the novel mutation operator. Experiments showed, the CSSMA outperforms than some classical algorithms in term of the detective rate and time cost.
Aiming at solving the problem of "self-set incomplete" that exists in intrusion detection system based on computer immune, this paper designs a construction algorithm of principal and subordinate structure s...
详细信息
Aiming at solving the problem of "self-set incomplete" that exists in intrusion detection system based on computer immune, this paper designs a construction algorithm of principal and subordinate structure self-set based on decision trees. The decision trees are introduced to traditional negative-selectionalgorithm and the candidate detectors which have been eliminated by the immune tolerance are reclassified by the decision trees, and the candidate detectors that meet the setting conditions compose the "subordinate self-set" so as to achieve the dynamic expansion of the self-set. The unqualified elements in "subordinate self-set" are eliminated according to the "conflict of matching" method. Experimental results show that this algorithm is effective and can improve the recognition performance of the detectors.
In this paper we first propose a novel neural networks-based negative selection algorithm (NSA). The principle and structure of our NSA are presented, and its training algorithm is derived Taking advantage of neural n...
详细信息
ISBN:
(纸本)0780385667
In this paper we first propose a novel neural networks-based negative selection algorithm (NSA). The principle and structure of our NSA are presented, and its training algorithm is derived Taking advantage of neural networks training, it has the distinguished capability of adaptation, which is well suited for dealing with practical problems under time-varying circumstances. A new fault diagnosis scheme using this NSA is next introduced, Two illustrative simulations of anomaly detection in chaotic time series and inner raceway fault diagnosis of bearings demonstrate the efficiency of the proposed neural networks-based NSA.
With the increasing opening of Chinese Financial Market to the world, the international venture capital, especially the hedge fund, will make big profits through various operational means in this faulty environment. T...
详细信息
ISBN:
(纸本)0769528759
With the increasing opening of Chinese Financial Market to the world, the international venture capital, especially the hedge fund, will make big profits through various operational means in this faulty environment. To ensure the stability of stock market, a newly method should be urgently proposed to participate in detecting abnormity of stock market. This paper introduces an artificial immune method establishes the immune detecting system framework by mining the features of single stock's abnormal fluctuation and matching with macroeconomic indexes, uses negative selection algorithm recognizing "Self" and "Non-self" at the same time, while builds up detectors with alarm mechanism, providing a new idea for the management of stock risk.
暂无评论