摘要
互联网应用系统已成为人们获取信息的平台,用户每;天与应用系统进行交互,产生大量的行为数据,为了使根据用户行为数据进行推荐的手段满足更多的推荐场景,解决信息过载的问题,从海量的信息数据中挖掘用户的兴趣点并给用户进行推送,以用户上网行为特征的推荐算法研究成为学术领域热点。依据用户实时签到产生的地理位置信息和用户行为上下文特征研究协同过滤算法,可以构建出一个在线上线下相结合的互联网领域的应用方案,并进行实验验证,有较好的准确度。
Internet application system has become a platform for people to obtain information.Users interact with the application system every day and produce a large amount of behavior data.In order to satisfy more recommendation scenarios by means of recommendation based on user behavior data,solve the problem of information overload,mine users'interest points from massive information data and push users,the research of recommendation algorithm based on user's online behavior characteristics has become a hot topic in the academic field.By researching collaborative filtering algorithm based on the geographic location information and user behavior context features generated by real-time user check-in,an online and offline application scheme can be constructed in the Internet field,and experimental verification is also done,with better accuracy.
作者
周翔宇
高仲合
ZHOU Xiang-yu;GAO Zhong-he(College of Software,Qufu Normal University,Qufu Shandong 273100,China)
出处
《通信技术》
2019年第8期1928-1931,共4页
Communications Technology
基金
国家自然科学基金青年项目(No.61601261)
山东省自然科学基金博土基金(No.ZR2016FB20)
山东省高等学校科技计划(No.J17KA062)
教育部产学合作协同育人项目(No.201602028014)~~
关键词
推荐算法
用户行为
地理位置融合
用户相似度
recommendation algorithm
user behavior
geographic location fusion
user similarity
