The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Inspirés du comportement naturel d’autogestion du corps humain, les systèmes autonomes promettent d’injecter un comportement d’autogestion dans les systèmes logiciels. Un tel comportement permet des capacités d'auto-configuration, d'auto-réparation, d'auto-optimisation et d'auto-protection dans les systèmes logiciels. L'auto-configuration est requise dans les systèmes où l'efficacité est la question clé, tels que les environnements d'exécution en temps réel. Pour résoudre les problèmes d'autoconfiguration dans les systèmes autonomes, l'utilisation de diverses techniques de résolution de problèmes a été rapportée dans la littérature, notamment le raisonnement basé sur des cas. L'approche de raisonnement basée sur des cas exploite l'expérience passée qui peut être utile pour atteindre les capacités autonomes. Le processus d'apprentissage s'améliore à mesure que davantage d'expérience est ajoutée à la base de cas sous forme de cas. Il en résulte une base de cas plus large. Une base de cas plus grande réduit l’efficacité en termes de coût de calcul. Pour surmonter ce problème d'efficacité, cet article suggère de regrouper la base de cas, afin de trouver ensuite la solution au problème signalé. Cette approche réduit la complexité de la recherche en limitant un nouveau cas à un cluster pertinent dans la base de cas. Le regroupement de la base de cas est un processus unique et n'a pas besoin d'être répété régulièrement. L'approche proposée présentée dans cet article a été décrite sous la forme d'un nouveau cadre de RBC groupé. Le cadre proposé a été évalué sur une simulation d’application autonome de feux de forêt (AFFA). Cet article présente un aperçu de l'AFFA simulé et des résultats sur trois algorithmes de clustering différents pour regrouper la base de cas dans le cadre proposé. La comparaison des performances de l'approche CBR conventionnelle et de l'approche CBR groupée a été présentée en termes d'exactitude, de rappel et de précision (ARP) et d'efficacité de calcul.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Malik Jahan KHAN, Mian Muhammad AWAIS, Shafay SHAMAIL, "Improving Efficiency of Self-Configurable Autonomic Systems Using Clustered CBR Approach" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 11, pp. 3005-3016, November 2010, doi: 10.1587/transinf.E93.D.3005.
Abstract: Inspired from natural self-managing behavior of the human body, autonomic systems promise to inject self-managing behavior in software systems. Such behavior enables self-configuration, self-healing, self-optimization and self-protection capabilities in software systems. Self-configuration is required in systems where efficiency is the key issue, such as real time execution environments. To solve self-configuration problems in autonomic systems, the use of various problem-solving techniques has been reported in the literature including case-based reasoning. The case-based reasoning approach exploits past experience that can be helpful in achieving autonomic capabilities. The learning process improves as more experience is added in the case-base in the form of cases. This results in a larger case-base. A larger case-base reduces the efficiency in terms of computational cost. To overcome this efficiency problem, this paper suggests to cluster the case-base, subsequent to find the solution of the reported problem. This approach reduces the search complexity by confining a new case to a relevant cluster in the case-base. Clustering the case-base is a one-time process and does not need to be repeated regularly. The proposed approach presented in this paper has been outlined in the form of a new clustered CBR framework. The proposed framework has been evaluated on a simulation of Autonomic Forest Fire Application (AFFA). This paper presents an outline of the simulated AFFA and results on three different clustering algorithms for clustering the case-base in the proposed framework. The comparison of performance of the conventional CBR approach and clustered CBR approach has been presented in terms of their Accuracy, Recall and Precision (ARP) and computational efficiency.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.3005/_p
Copier
@ARTICLE{e93-d_11_3005,
author={Malik Jahan KHAN, Mian Muhammad AWAIS, Shafay SHAMAIL, },
journal={IEICE TRANSACTIONS on Information},
title={Improving Efficiency of Self-Configurable Autonomic Systems Using Clustered CBR Approach},
year={2010},
volume={E93-D},
number={11},
pages={3005-3016},
abstract={Inspired from natural self-managing behavior of the human body, autonomic systems promise to inject self-managing behavior in software systems. Such behavior enables self-configuration, self-healing, self-optimization and self-protection capabilities in software systems. Self-configuration is required in systems where efficiency is the key issue, such as real time execution environments. To solve self-configuration problems in autonomic systems, the use of various problem-solving techniques has been reported in the literature including case-based reasoning. The case-based reasoning approach exploits past experience that can be helpful in achieving autonomic capabilities. The learning process improves as more experience is added in the case-base in the form of cases. This results in a larger case-base. A larger case-base reduces the efficiency in terms of computational cost. To overcome this efficiency problem, this paper suggests to cluster the case-base, subsequent to find the solution of the reported problem. This approach reduces the search complexity by confining a new case to a relevant cluster in the case-base. Clustering the case-base is a one-time process and does not need to be repeated regularly. The proposed approach presented in this paper has been outlined in the form of a new clustered CBR framework. The proposed framework has been evaluated on a simulation of Autonomic Forest Fire Application (AFFA). This paper presents an outline of the simulated AFFA and results on three different clustering algorithms for clustering the case-base in the proposed framework. The comparison of performance of the conventional CBR approach and clustered CBR approach has been presented in terms of their Accuracy, Recall and Precision (ARP) and computational efficiency.},
keywords={},
doi={10.1587/transinf.E93.D.3005},
ISSN={1745-1361},
month={November},}
Copier
TY - JOUR
TI - Improving Efficiency of Self-Configurable Autonomic Systems Using Clustered CBR Approach
T2 - IEICE TRANSACTIONS on Information
SP - 3005
EP - 3016
AU - Malik Jahan KHAN
AU - Mian Muhammad AWAIS
AU - Shafay SHAMAIL
PY - 2010
DO - 10.1587/transinf.E93.D.3005
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2010
AB - Inspired from natural self-managing behavior of the human body, autonomic systems promise to inject self-managing behavior in software systems. Such behavior enables self-configuration, self-healing, self-optimization and self-protection capabilities in software systems. Self-configuration is required in systems where efficiency is the key issue, such as real time execution environments. To solve self-configuration problems in autonomic systems, the use of various problem-solving techniques has been reported in the literature including case-based reasoning. The case-based reasoning approach exploits past experience that can be helpful in achieving autonomic capabilities. The learning process improves as more experience is added in the case-base in the form of cases. This results in a larger case-base. A larger case-base reduces the efficiency in terms of computational cost. To overcome this efficiency problem, this paper suggests to cluster the case-base, subsequent to find the solution of the reported problem. This approach reduces the search complexity by confining a new case to a relevant cluster in the case-base. Clustering the case-base is a one-time process and does not need to be repeated regularly. The proposed approach presented in this paper has been outlined in the form of a new clustered CBR framework. The proposed framework has been evaluated on a simulation of Autonomic Forest Fire Application (AFFA). This paper presents an outline of the simulated AFFA and results on three different clustering algorithms for clustering the case-base in the proposed framework. The comparison of performance of the conventional CBR approach and clustered CBR approach has been presented in terms of their Accuracy, Recall and Precision (ARP) and computational efficiency.
ER -