论文部分内容阅读
针对标准人工蜂群算法搜索效率低、收敛速度慢等缺点提出一种改进的人工蜂群算法.通过引入算术交叉操作以及利用最优解指导搜索方向,增加算法收敛的速度.在7个基准函数上的测试结果表明了算法的有效性.在此基础上,针对K-means算法的缺点提出基于改进蜂群算法的K-means算法,并加入自动获得最佳聚类数的功能.在人工数据集和UCI真实数据集上的测试验证了所提出算法的性能.
This paper proposes an improved artificial bee colony algorithm based on the low search efficiency and slow convergence speed of standard artificial bee colony algorithm.The algorithm accelerates the speed of convergence by introducing arithmetic crossover and using the optimal solution to guide the search direction.In the seven benchmark functions On the basis of which the K-means algorithm based on the improved bee colony algorithm is proposed for the shortcomings of K-means algorithm, and the function of automatically obtaining the best clustering number is added in the test results on the artificial data The tests on the set and UCI real data sets verify the performance of the proposed algorithm.