Jason Brownlee · Swinburne Research Bank (Swinburne University of Technology) · 2007
The clonal selection principle describes the selection and maturation of lymphocytes, accounting for the diversity of the antibodies they produce and the tolerance they have to selfantigens. The theory does not exist in isolation, the receptors, the host, and the antigens have a context that is commonly ignored when this principle is realised as a computational intelligence algorithm. This work attempts to provide context to the principle by introducing a spatial structure in which to house membrane bound receptors, and which simple clonal selection operations (as previously defined for the elaborated clonal selection algorithm) are applied.
Cited by 1Open access
Yong Peng, Bao‐Liang Lu · Information Sciences · 2014
Cited by 34
Anurag Sharma, Dharmendra Sharma · Lecture notes in computer science · 2011
Cited by 34
Jason Brownlee · Swinburne Research Bank (Swinburne University of Technology) · 2005
The vertebrate immune system is a robust and powerful information process system that demonstrates features such as distributed control, parallel processing and adaptation and learning via experience. Artificial Immune Systems (AIS) are machine-learning algorithms that embody some of the principles and attempt to take advantages of the benefits of natural immune systems for use in tackling complex problem domains. The Clonal Selection Algorithm (CLONALG) is one such system inspired by the clonal selection theory of acquired immunity which has shown success on broad range of engineering problem domains. The focus of this work is the CLONALG algorithm specifically the techniques history previous research and algorithm function. In addition this work seeks to take desirable elements from CLONALG implementations and devise a clonal selection based classification algorithm. Competence with the CLONALG algorithm is demonstrated in terms of theory and application. The algorithm is analysed from the perspective of desirable features for an immune inspired classifier and a number a new clonal selection inspired technique called the Clonal Selection Classification Algorithm (CSCA) is proposed designed specified and preliminary tested. The technique is shown to be robust self-tuning (in terms of resources used) insensitive to user parameters and competitive as a classification system on common problem domains. An implementation of the CSCA is provided as a plug-in for the WEKA machinelearning workbench as is a naïve implementation of the CLONALG algorithm for classification.
Cited by 39Open access
Jason Brownlee · Swinburne Research Bank (Swinburne University of Technology) · 2007
The clonal selection theory suggests an information processing principle of the selection and proliferation of temporally useful information. The proliferation strategy promotes maturation of response via blind mutation and anticipation of selection through adjusted information densities. This work explores the clonal selection principle as an information-processing algorithm and proposes (1) a minimal approach, (2) an elaborated approach, and (3) implications and extensions of the proposed system.
Cited by 9Open access
Jacek Dąbrowski · 2008
Over the years several successful computing techniques have been inspired by biological mechanisms. Studies of the mechanisms that allow the immune systems of vertebrates to adapt and learn have resulted in a class of algorithms called artificial immune systems. Clonal selection is a process that allows lymphocytes to launch a quick response to known pathogens and to adapt to new, previously un-encountered ones. This paper presents an algorithm based on the clonal selection principles for solving the capacitated vehicle routing problem.
Cited by 5
Johnny Koh Siaw Paw, Chong Kok Hen, Koo Wai Yan et al. · 2012
In this paper, the performance of a clonal selection optimization algorithm implementing a combined mutation strategy using Gaussian and Cauchy mutations was further improved with the addition of two new operators, namely the Cell Repair Operator (CPO) and Dynamic Mutation Size Operator (DMSO). The new and improved algorithm was tested using various unimodal and multimodal test functions to illustrate the improvements of the algorithm. The test results indicate that the added operators have significantly improved the performance of the algorithm in terms of accuracy and performance.
Cited by 0
Ruochen Liu, Xiangrong Zhang, Neng Yang et al. · Applied Soft Computing · 2011
Cited by 22
Yi-Hui Su, Wen‐Jye Shyr, Te‐Jen Su · Lecture notes in computer science · 2005
Cited by 14
Jason Brownlee, Jason Brownlee · 2005
The vertebrate immune system is a robust and powerful information process system that demonstrates features such as distributed control, parallel processing and adaptation and learning via experience. Artificial Immune Systems (AIS) are machine-learning algorithms that embody some of the principles and attempt to take advantages of the benefits of natural immune systems for use in tackling complex problem domains. The Clonal Selection Algorithm (CLONALG) is one such system inspired by the clonal selection theory of acquired immunity, which has shown success on broad range of engineering problem domains. The focus of this work is the CLONALG algorithm, specifically the techniques history, previous research and algorithm function. In addition, this work seeks to take desirable elements from CLONALG implementations and devise a clonal selection based classification algorithm. Competence with the CLONALG algorithm is demonstrated in terms of theory and application. The algorithm is analysed from the perspective of desirable features for an immune inspired classifier and a number a new clonal selection inspired technique called the Clonal Selection Classification
Cited by 0
Simone A. Ludwig · Zenodo (CERN European Organization for Nuclear Research) · 2012
Quality of Service (QoS) aware service selection of workflows is a very important aspect for service-oriented systems. The selection based on QoS allows the user to include also non-functional attributes in their query, such as availability and reliability. Several exact methods have been proposed in the past, however, given that the workflow selection problem is NP-hard, approximate algorithms can be used to find suboptimal solutions for requested workflows. Genetic algorithm is one such method that can find approximate solutions in the form of services selected. In this paper, we propose an improved version of the standard genetic algorithm approach by making use of the clonal selection principle from artificial immune systems. Experimental results show that the clonal selection based genetic algorithm achieves much higher fitness values for the workflow selection problem than standard genetic algorithm.
Cited by 23Open access
Ayi Purbasari, DS · Universitas Pasundan institutional repositories & scientific journals (Universitas Pasundan) · 2016
Cited by 0
Lihua Zhao, Jin Pan · International Journal of Modelling Identification and Control · 2013
In this paper, we present an directed clonal selection algorithm (DCLONALG) for mining association rules. Different from the traditional evolutionary algorithms, DCLONALG firstly scans dataset one time to find frequent rules with only one item. Then, these items were used to generate new rules and the mutation operation was limited in it. We evaluate the performance of the proposed DCLONALG approach for AC and the performance results have shown that the proposed approach is efficient in dealing with the problem on the complexity of the rule search space. At the same time, good classification accuracy has been achieved. This is especially important for mining association rules from large datasets in which the search space of rules is huge.
Cited by 3
Yaowen Chen, Lin Huang, Weiming Luo et al. · 2008
According to the basis of clonal selection immune algorithm and hierarchical clustering, a dynamic clonal selection immune clustering algorithm is presented, which no pre-knowledge is needed. The proposed algorithm bases on antibody affinity, to recognize antigen, restrain and merge antibody. By using aiNET immune network model, the algorithm mutates location of antibodies, in which the mutating rate is dynamically adjusted with inverse proportion to the number of immune evolution generations. After dynamic mutation, the similar antibodies are merged again, and the same processes repeats until it meets the ending condition. Experimental results showed that the proposed algorithm is more coincidental reality of clustering and more preferable performance than traditional ones.
Cited by 2
Hongwei Dai, Yu Yang, Hui Li et al. · International Journal of Advancements in Computing Technology · 2011
Clonal selection theory is one of the important theories in immunological area. Applying clonal selection to optimization problems can also be found in many papers. In this paper, the classical Clonal Selection Algorithm (CSA) is extended by introducing the receptor editing mechanism which provides a chance for autoreactive cells to survive during affinity maturation process. To demonstrate the effectiveness and applicability of the novel CSA (NCSA), experiments are carried out on the knapsack problem. Simulation results show that the NCSA has a better performance when compared to the traditional CSA and other algorithms.
Cited by 1
Lining Zhang, Maoguo Gong, Licheng Jiao et al. · Journal of Software · 2010
基于免疫学中的抗体克隆选择学说,通过引入抗独特型结构,提出了一种用于求解复杂多峰函数优化问题人工免疫系统算法——抗独特型克隆选择算法.该算法通过克隆增殖操作、抗独特型变异操作、抗独特型重组操作和克隆选择操作这4 个操作算子来实现抗体种群的进化,能够同时在同一抗体周围的多个方向进行全局搜索和局部搜索,具有较强的搜索能力.理论分析表明,抗独特型克隆选择算法具有全局收敛性.抗独特型结构的引入充分利用了优势抗体的结构信息,加快了抗体种群的收敛速度,从而以更快的速度获得全局最优解,同时降低了算法陷入局部极值点的几率.实验部分采用4 组不同类型的函数对算法性能进行测试.理论分析及实验结果表明,与克隆选择算法等已有算法相比,该算法性能好,求解精度高,鲁棒性强.
Cited by 1Open access
Margarita-Arimatea Díaz-Cortés, Erik Cuevas, Raúl Rojas · Intelligent systems reference library · 2017
Cited by 4
Hongwei Dai, Yang Yu, Hui Li et al. · International Journal of Digital Content Technology and its Applications · 2012
Cited by 0
Xianhua Zheng, Luo Yanmin · Journal of Computer Applications · 2013
Cited by 0
Leandro Nunes de Castro, Fernando J. Von Zuben · 2007
The clonal selection algorithm is used by the natural immune system to define the basic features of an immune response to an antigenic stimulus. It establishes the idea that only those cells that recognize the antigens are selected to proliferate. The selected cells are subject to an affinity maturation process, which improves their affinity to the selective antigens. In this paper, we propose a powerful computational implementation of the clonal selection principle that explicitly takes into account the affinity maturation of the immune response. The algorithm is shown to be an evolutionary strategy capable of solving complex machinelearning tasks, like pattern recognition and multimodal optimization. 1 INTRODUCTION Over the last few years, there has been an ever increasing interest in the area of artificial immune systems (AIS) and their applications. Among the many works in this new field of research, we can detach those of Ishida (1996); Hunt & Cook (1996); Dasgupt...
Cited by 536