基于DBS-PSO优化算法在关联规则挖掘中的研究与应用
发布时间：2018-10-12 21:39
【摘要】：关联规则挖掘是数据挖掘技术领域内的研究重点和热点之一,在各行业领域内有着广泛的应用,Apriori算法作为关联规则的代表性算法之一,其性能的好坏直接关系到关联分析的效率和结论。目前,面对爆炸式增长的各类数据,Apriori算法在处理时,其面临问题也日益突出,主要体现在算法运行时间长、效率低以及需要通过主观单一设置最小支持度和最小置信度的阈值实现关联规则提取这两方面上。近年来,国内外相关学者都对Apriori算法的改进优化进行了研究,其中,将Apriori算法与其它智能算法融合进行改进是当前一个研究热点,且在此研究方向上取得了丰硕的研究成果。结合以上情况,本文提出了一种DBS-PSO优化算法对进行优化研究,其基本思路是:首先,利用改进的密度偏差抽样算法对原始数据集进行抽样,获取样本数据;其次,通过设置适应度函数,利用改进的粒子群算法迭代寻优获取解空间;最后,将粒子群算法求解的解空间作为Apriori算法中最小支持度和置信度的阈值,对样本数据进行关联规则挖掘。实验仿真结果表明:在Apriori算法的优化研究上,本文提出的DBS-PSO优化算法,不仅降低了 Apriori算法的时间运行成本,同时使得关联规则的挖掘更为合理、客观和高效。
[Abstract]:Association rule mining is one of the research focuses and hotspots in the field of data mining technology. It has been widely used in various fields. Apriori algorithm is one of the representative algorithms of association rules. Its performance is directly related to the efficiency and conclusion of correlation analysis. At present, in the face of explosive growth of all kinds of data, Apriori algorithm in processing, its problems are increasingly prominent, mainly reflected in the algorithm running time long, Low efficiency and the need to set a single subjective minimum support and the minimum confidence of the threshold to achieve association rules extraction. In recent years, scholars at home and abroad have studied the improved optimization of Apriori algorithm. Among them, improving the fusion of Apriori algorithm and other intelligent algorithms is a research hotspot, and has achieved a lot of research results in this research direction. Combined with the above situation, this paper proposes a DBS-PSO optimization algorithm for optimization research. The basic ideas are: firstly, the improved density deviation sampling algorithm is used to sample the original data set to obtain the sample data. By setting the fitness function, the improved particle swarm optimization algorithm is used to find the solution space. Finally, the solution space of the particle swarm optimization algorithm is used as the threshold of the minimum support and confidence in the Apriori algorithm. Mining association rules for sample data. The simulation results show that the DBS-PSO optimization algorithm proposed in this paper not only reduces the time cost of Apriori algorithm, but also makes the mining of association rules more reasonable, objective and efficient.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13
[Abstract]:Association rule mining is one of the research focuses and hotspots in the field of data mining technology. It has been widely used in various fields. Apriori algorithm is one of the representative algorithms of association rules. Its performance is directly related to the efficiency and conclusion of correlation analysis. At present, in the face of explosive growth of all kinds of data, Apriori algorithm in processing, its problems are increasingly prominent, mainly reflected in the algorithm running time long, Low efficiency and the need to set a single subjective minimum support and the minimum confidence of the threshold to achieve association rules extraction. In recent years, scholars at home and abroad have studied the improved optimization of Apriori algorithm. Among them, improving the fusion of Apriori algorithm and other intelligent algorithms is a research hotspot, and has achieved a lot of research results in this research direction. Combined with the above situation, this paper proposes a DBS-PSO optimization algorithm for optimization research. The basic ideas are: firstly, the improved density deviation sampling algorithm is used to sample the original data set to obtain the sample data. By setting the fitness function, the improved particle swarm optimization algorithm is used to find the solution space. Finally, the solution space of the particle swarm optimization algorithm is used as the threshold of the minimum support and confidence in the Apriori algorithm. Mining association rules for sample data. The simulation results show that the DBS-PSO optimization algorithm proposed in this paper not only reduces the time cost of Apriori algorithm, but also makes the mining of association rules more reasonable, objective and efficient.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13
【参考文献】
相关期刊论文 前10条
1 韩家琪;毛克彪;夏浪;刘R，
本文编号：2267688
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2267688.html
