作业帮 > 英语 > 作业

英语翻译Creating individual consumer profiles for personal

来源:学生作业帮 编辑:搜搜做题作业网作业帮 分类:英语作业 时间:2024/04/29 22:23:33
英语翻译
Creating individual consumer profiles for personalized recommendation (or for other
purposes,such as providing dynamic content or tailored advertizing) exacerbates this prob-
lem,because now one may be searching for patterns individually for each of millions of
consumers.Adomavicius and Tuzhilin (2001) address this problem.They show how to
automate,partially,the process of expert-driven validation and filtering of large sets of
rules.Their method comprises various operators for browsing,grouping,validating,and
filtering rules.They demonstrate the method by applying it to data on consumer purchases
of beverages—about 2000 households over a year period.For example,association-rule
mining produced over one million rules from these data.In about an hour and a half,com-
prising mostly browsing and thinking,the expert-filtering process had rejected definitively
96% of the rules,and had used 27,000 rules to build individual profiles for the households
(averaging about 14 rules per profile).
英语翻译Creating individual consumer profiles for personal
感觉这个很像是外文文献的内容.
以下仅供参考:
为了个人化的推荐而创造个人消费者形象(或者其他用途,比如提供动态内容或量身定做的推广),使这个问题更严重了,因为现在人们可以单独地搜索成千上万的消费者中每个人的模式.
Adomavicius和Tuzhilin(2001)提出了这个问题.他们展示了如何使专家驱动(expert-driven)验证和对大量规则过滤的过程进行自动化.他们的方法包括了各种操作人员的浏览、分组、验证和过滤规则.他们通过把它应用到消费者购买酒水(大概2000户家庭一年多的时间)的数据上,来展现了这种方法.
比如,关联规则挖掘,从这些数据产生了超过一百万的规则.在大概一个半小时内,包括大部分浏览和思考,专家过滤过程中曾最终拒绝96%的规则,并使用了2.7万的规则来为这些家庭建立个人的档案(平均每个档案14条规则).