《計算機應用研究》|Application Research of Computers

融合社交行為和標簽行為的推薦算法研究

Study of recommended algorithm integrating social behavior and labeling behavior

免費全文下載 (已被下載 次)  
獲取PDF全文
作者 蔣云,倪靜,房宏揚
機構 上海理工大學 管理學院,上海 200093
統計 摘要被查看 次,已被下載
文章編號 1001-3695(2019)07-010-1965-05
DOI 10.19734/j.issn.1001-3695.2018.01.0038
摘要 針對傳統推薦算法忽略用戶社交影響、研究角度不全面和缺乏物理解釋等問題,提出一個融合社交行為和標簽行為的推薦算法。首先用引力模型計算社交網絡中用戶節點之間的吸引力來度量用戶社交行為的相似性;其次通過標簽信息構建用戶喜好物體模型,并使用引力公式計算喜好物體之間的引力來度量標簽行為的相似性。最后,引入變量融合兩方面信息,獲取近鄰用戶,產生推薦。采用Last.fm數據集進行實驗研究,結果說明推薦算法的準確率和召回率更高。
關鍵詞 社交行為; 標簽行為; 萬有引力; 協同過濾
基金項目 國家自然科學基金面上項目(71774111)
本文URL http://www.oirznw.live/article/01-2019-07-010.html
英文標題 Study of recommended algorithm integrating social behavior and labeling behavior
作者英文名 Jiang Yun, Ni Jing, Fang Hongyang
機構英文名 School of Business,University of Shanghai for Science & Technology,Shanghai 200093,China
英文摘要 In view of the traditional recommendation algorithm ignoring the impact of social behavior of users, the incomprehensive research perspective and lack of physical explanation, this paper proposed a recommendation algorithm that integrated social behavior and tagging behavior of users. Firstly, it calculated the attractiveness between user nodes in social network by gravity model to measure the similarity of users' social behavior. Secondly, it constructed the user's favorite object model by label information, also used the gravitation formula to calculate the gravitation between favorite objects to measure the similarity of tagging behavior. Finally, the paper introduced the variables to weigh the proportion of two similar values, and then got the set of neighbors and generated recommendations. Experimental results using Last. fm dataset show that the proposed algorithm has higher precision and recall.
英文關鍵詞 social behavior; labeling behavior; gravitation; collaborative filtering
參考文獻 查看稿件參考文獻
 
收稿日期 2018/1/22
修回日期 2018/3/6
頁碼 1965-1969
中圖分類號 TP301.6
文獻標志碼 A
水果机返水