摘要
为解决电子商务中存在的用户搜索商品耗时太长的问题,提出了利用交互式遗传算法的个性化商品搜索。根据用户输入的初始搜索字段,利用基于关键词的多层递阶编码结构,构造进化个体,并给出此类编码方式的解码、交叉、变异算子。用户选择的潜在感兴趣的商品信息,提出了用户行为的进化个体适应值评价模型,以获得用户对所有商品的感兴趣度,进而利用交互式遗传算法辅助用户尽快搜索到满意信息。最后,将改进算法应用于JADE平台开发的智能购书系统,通过与现有购物平台在搜索耗时和成功率方面的比较验证了本文方法的有效性。
In order to solve the problem that users spend a lot of time finding the satisfied goods in e - commerce, a personal search system was proposed by adopting an interactive genetic algorithm. Firstly, a multi - layer hierarchical structure was designed to encode the goods according to their key words and the ones entered by users, and the encoded key words were viewed as evolutionary individuals. With this novel encoding mechanism, corresponding methods of decoding, crossover and mutation were presented. For effectively comparing the satisfied degrees of users with all displayed goods, a fitness function was built based on the goods which users had evaluated and may potentially interested in. Lastly, an intelligent E -commerce shopping system with JADE was developed and compared with the traditional systems, and the results show that our algorithm is obviously advanced in saving user time and improving trade success.
出处
《计算机仿真》
CSCD
北大核心
2013年第2期295-298,共4页
Computer Simulation
基金
江苏省科技计划项目(BC2010058)
中央高校基本科研业务费专项资金资助(2010QNB31)
关键词
电子商务
交互
遗传算法
递阶编码
E - commerce
Interaction
Genetic algorithm
Hierarchical encoding
