安徽理工大学学报(自然科学版)
安徽理工大學學報(自然科學版)
안휘리공대학학보(자연과학판)
JOURNAL OF ANHUI UNIVERSITY OF SCIENCE AND TECHNOLOGY(NATURAL SCIENCE)
2015年
1期
55-59,82
,共6页
UMARS%用户模型%空间扩展%属性扩散%时间更新
UMARS%用戶模型%空間擴展%屬性擴散%時間更新
UMARS%용호모형%공간확전%속성확산%시간경신
UMARS%user model%spatial diffusion%attribute diffusion%time update
电子商务的飞速发展,在为人们提供了更加方便快捷的购物途径同时,如何给用户提供更加人性化的推荐服务,是电商赢得市场需要解决的关键问题。以淘宝电子商务平台为依托,提出基于用户模型和商品属性扩散的混合个性化推荐系统,研究用户兴趣的时间演化和空间扩散方法,综合运用多种推荐方法,给为用户提供更加精准的推荐服务。实验证明,该方法具有较明显的效果,能够较好地运用于同类电商平台。
電子商務的飛速髮展,在為人們提供瞭更加方便快捷的購物途徑同時,如何給用戶提供更加人性化的推薦服務,是電商贏得市場需要解決的關鍵問題。以淘寶電子商務平檯為依託,提齣基于用戶模型和商品屬性擴散的混閤箇性化推薦繫統,研究用戶興趣的時間縯化和空間擴散方法,綜閤運用多種推薦方法,給為用戶提供更加精準的推薦服務。實驗證明,該方法具有較明顯的效果,能夠較好地運用于同類電商平檯。
전자상무적비속발전,재위인문제공료경가방편쾌첩적구물도경동시,여하급용호제공경가인성화적추천복무,시전상영득시장수요해결적관건문제。이도보전자상무평태위의탁,제출기우용호모형화상품속성확산적혼합개성화추천계통,연구용호흥취적시간연화화공간확산방법,종합운용다충추천방법,급위용호제공경가정준적추천복무。실험증명,해방법구유교명현적효과,능구교호지운용우동류전상평태。
With the rapid development of e -commerce, which provides convenient and efficient way for shop-ping, how to provide users with more personalized recommendation service , is the key problem to be solved by sellers.Taking Taobao , an e-commerce platform as studying object , hybrid personalized recommendation sys-tem based on the user model and product attributes diffusion was proposed .Based on research of user interest time evolution and space diffusion method , by using various recommended methods , a more accurate recommen-dation was provided to users .Experimental results showed that the method is effective and can be used in similar business platform .