转换参数非线性递减的正弦余弦算法

发布时间：2019-06-02 00:48

【摘要】：正弦余弦算法是一种新型智能优化算法,利用正弦函数和余弦函数值的变化来实现优化搜索。转换参数直接影响算法全局探索和局部开发的平衡,对算法的性能有着重要影响。为提高该算法的优化性能,首先对转换参数的设置进行分析,然后设计出转换参数抛物线函数递减和指数函数递减两种正弦余弦算法,并采用标准测试函数进行数值实验,和转换参数线性递减的基本正弦余弦算法进行比较。结果表明指数函数递减的正弦余弦算法具有更高的计算精度和更快的收敛速度。最后以协同过滤推荐算法中相似度函数的计算为应用对象,进一步验证新算法的可行性和有效性。
[Abstract]:Sinusoidal cosine algorithm is a new intelligent optimization algorithm, which uses the change of sinusoidal function and cosine function value to realize the optimization search. The conversion parameters directly affect the balance between global exploration and local development of the algorithm, and have an important impact on the performance of the algorithm. In order to improve the optimization performance of the algorithm, the setting of conversion parameters is analyzed, and then two sinusoidal cosine algorithms, parabola function decline and exponential function decline, are designed, and the standard test function is used to carry out numerical experiments. It is compared with the basic sinusoidal cosine algorithm with linear decline of conversion parameters. The results show that the sinusoidal cosine algorithm with decreasing exponential function has higher calculation accuracy and faster convergence speed. Finally, the calculation of similarity function in collaborative filtering recommendation algorithm is taken as the application object to further verify the feasibility and effectiveness of the new algorithm.
【作者单位】：上海理工大学管理学院;
【基金】：国家自然科学基金(No.71401106) 教育部人文社会科学研究规划基金(No.16YJA630037) 上海市高原学科建设项目上海高校青年教师培养资助计划项目(No.ZZsl15018) 上海理工大学国家级培育青年基金(No.16HJPY-QN15);上海理工大学博士科研启动经费项目(No.1D-15-303-005)
【分类号】：TP18

【相似文献】