摘要
针对标准K-means算法在对教学数据分析时还存在效果不好等问题,本文提出一种基于种群优化遗传算法优化K-means聚类的高校艺术教学数据分析模型,首先为了保证遗传算法的种群多样性,然后利用小生境方法限制种群个体的繁衍,以达到种族多样化的优化,接着定义了染色体的惩罚函数,根据染色体对数据的聚类结果来动态调节染色体的适合度,防止染色体早熟现象的出现,最后构建高校艺术教学数据分析模型。仿真试验结果表明,基于种群优化遗传算法优化K-means聚类的高校艺术教学数据分析模型相比较标准K-means算法具有更好的分析效果。
In view of the bad performance of standard K-means algorithm in the teaching data analysis, this paper put forward a teaching data analysis and management model of college arts education based on genetic algorithm optimized K-means clustering. Firstly, in order to keep the population variety of genetic algorithm, niche method is used to limit the reproduction of individuals and optimize the population variety. Then the punishment function of chromosome is defined. The fitness of chromosome is dynamically adjusted according to the clustering results of chromosome data so as to avoid the premature of chromosome. Finally, the college arts education data analysis model is established. The simulation experiments show that the data analysis model based on K-means clustering optimized by population optimization genetic algorithm has better performance than standard K-means algorithm.
出处
《科技通报》
北大核心
2016年第4期231-234,共4页
Bulletin of Science and Technology
基金
2014年江西省高校人文社科立项课题<高职院校艺术教育特色再创性提升研究>(课题编号:YS1431)
