摘要
企业为了扩大销售,就必须根据市场细分的原则,通过对消费者的聚类,了解不同消费者的需要。传统的K均值聚类分析对于初始聚类的中心点比较敏感,借助于遗传算法能够搜索到全局最优解的特点,可以克服传统方法的缺点。运用基于GA的K均值聚类分析方法,对于86个不愿意购买某品牌方便面的大学生进行调查,结果表明男生主要是对该方便面的外观包装和面饼大小不满意,女生主要是对它的口感和外观包装不满意。这一分析结果对于企业改进产品设计是有帮助的。
For the purpose of extending distribution, enterprise must know the customer needs by the clustering based on market subdivision. Traditional K-means clustering analysis is sensitive to the initial central points. Due to the advantage of global search ability of genetic algorithm ,which can overcome the shortage of the traditional cluster method. The GA-based K-means clustering analysis was used to treat the data from 86 undergraduate who were unwilling to buy a certain brand instant noodle. The results showed males were dissatisfied with the appearance and quantity of noodle ,while females were dissatisfied with the flavor and appearance. It will be helpful for the enterprise to improve the product design.
出处
《广西师范大学学报（自然科学版）》
CAS
北大核心
2009年第3期81-84,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
国家社会科学基金"十一五"规划资助项目(BBA080050)
关键词
遗传算法
K均值聚类分析
市场调查
消费者心理
genetic algorithm
K-means clustering analysis
marketing research
customer psychology
