Trends in algorithm design have shown that hybrid algorithms, which combine or merge multiple algorithms, can create synergies to overcome the inherent limitations of the underlying individual algorithms. There are tw...
详细信息
Trends in algorithm design have shown that hybrid algorithms, which combine or merge multiple algorithms, can create synergies to overcome the inherent limitations of the underlying individual algorithms. There are two broad types of hybridization: collaborative - individual algorithms tackle an instance of the problem sequentially or in parallel and exchange information accordingly while solving the problem; integrative - individual algorithms are dedicated to tackling different aspect(s) of the problem-solving process. In this thesis, we propose a schema for an enhanced form of integrative hybridization that blends granular algorithmic components from multiple algorithms to derive a new singular clusteringalgorithm. As a case study for the proposed hybridization technique, we examine ant clustering algorithm (a swarm intelligence algorithm that is based on the natural phenomenon of brood sorting in some species of ants); highlight the strengths and weaknesses of the algorithm; and present a blend of algorithmic components from tabu search into the algorithm to improve its solution quality. Empirical results from applying the blended algorithm to clustering benchmark datasets show improved clustering validation measures for the proposed blended hybrid algorithm compared to other forms of hybridization of the same underlying individual algorithms. Besides, the quality of clusters uncovered by this hybrid algorithm competes favorably with those uncovered using popular clusteringalgorithms such as DBSCAN and mean shift. Finally, we show the feasibility and viability of the blended algorithm when used in a novel application of clustering to the estimation of the cost of claims for group insurance benefits.
Considering the existing massive volumes of data processed nowadays and the distributed nature of many organizations, there is no doubt how vital the need is for distributed database systems. In such systems, the resp...
详细信息
Considering the existing massive volumes of data processed nowadays and the distributed nature of many organizations, there is no doubt how vital the need is for distributed database systems. In such systems, the response time to a transaction or a query is highly affected by the distribution design of the database system, particularly its methods for fragmentation, replication, and allocation data. According to the relevant literature, from the two approaches to fragmentation, namely horizontal and vertical fragmentation, the latter requires the use of heuristic methods due to it being NP-Hard. Currently, there are a number of different methods of providing vertical fragmentation, which normally introduce a relatively high computational complexity or do not yield optimal results, particularly for large-scale problems. In this paper, because of their distributed and scalable nature, we apply swarm intelligence algorithms to present an algorithm for finding a solution to vertical fragmentation problem, which is optimal in most cases. In our proposed algorithm, the relations are tried to be fragmented in such a way so as not only to make transaction processing at each site as much localized as possible, but also to reduce the costs of operations. Moreover, we report on the experimental results of comparing our algorithm with several other similar algorithms to show that ours outperforms the other algorithms and is able to generate a better solution in terms of the optimality of results and computational complexity.
clustering analysis is used in many disciplines and applications. This data mining task is an essential tool that identifies groups of objects based on similarity measures. The ant clustering algorithm is a swarm-inte...
详细信息
clustering analysis is used in many disciplines and applications. This data mining task is an essential tool that identifies groups of objects based on similarity measures. The ant clustering algorithm is a swarm-intelligent method used for clustering problems, that is inspired by the natural behavior of ants, which collect their corpses and sort their larvae. In recent years, many new metaheuristic algorithms have explored different scientific fields using inspirations or motivations in biological or nature-inspired derivations. The same situation is observed in area of data mining, where plenty of metaheuristics was presented as a new and effective approach. This state causes some difficulties or disappointments to have enough explanations or deep extent analysis to justify such applications. A new version of an ant clustering algorithm, using cellular automata mechanisms of transitions, is proposed to examine the computational efficiency and accuracy of the antclustering approach. Based on the similarity of two approaches arising from ant Sleeping Model, computational analysis of difficulties and complexities are presented and results show that the modification of an ant clustering algorithm - ASM MOD produces results that are not only more stable but also more efficiently determined than the prototype of ASM. The aim of this paper is not explicitly criticize these approaches but also to find a way of the repair these algorithms.
In this paper we deal with the combinatorial state space explosion problem which can occur in the generation of the state space in explicit model checking. The generation phase will determine the performances of the r...
详细信息
ISBN:
(纸本)9781467396691
In this paper we deal with the combinatorial state space explosion problem which can occur in the generation of the state space in explicit model checking. The generation phase will determine the performances of the rest model checking, which justify the need of efficient methods to enhance the generated state space and obtain good solutions. We define the parallel generation of the state space as an optimization problem, and we propose an ant-based clusteringalgorithm to optimize the parallel generation in terms of balancing the workload and minimizing the communications overhead. Earlier experimental measures performed on cluster of machines have shown very promising results.
暂无评论