武汉理工大学学报(信息与管理工程版)
武漢理工大學學報(信息與管理工程版)
무한리공대학학보(신식여관리공정판)
JOURNAL OF WUHAN AUTOMOTIVE POLYTECHNIC UNIVERSITY
2014年
3期
365-368
,共4页
刘旭东%崔蕾%陈德人
劉旭東%崔蕾%陳德人
류욱동%최뢰%진덕인
协同过滤%选择性预测策略%平均绝对偏差
協同過濾%選擇性預測策略%平均絕對偏差
협동과려%선택성예측책략%평균절대편차
collaborative filtering%selective prediction strategy%mean absolute error
针对传统协同过滤推荐算法在用户评分数据极端稀疏情况下无法取得令人满意的推荐质量问题,结合User-based 和Item-based 协同过滤算法思想,提出了一种基于选择性预测策略的协同过滤推荐算法,算法利用高相似度阈值来计算用户相似性和项目相似性,并通过形成用户最近邻居集和项目最近邻居集来预测填充评分矩阵。基于Movielens数据集的实验表明,改进的算法有效改善了传统协同过滤推荐算法的数据稀疏性和扩展性问题,明显提高了系统的推荐质量。
針對傳統協同過濾推薦算法在用戶評分數據極耑稀疏情況下無法取得令人滿意的推薦質量問題,結閤User-based 和Item-based 協同過濾算法思想,提齣瞭一種基于選擇性預測策略的協同過濾推薦算法,算法利用高相似度閾值來計算用戶相似性和項目相似性,併通過形成用戶最近鄰居集和項目最近鄰居集來預測填充評分矩陣。基于Movielens數據集的實驗錶明,改進的算法有效改善瞭傳統協同過濾推薦算法的數據稀疏性和擴展性問題,明顯提高瞭繫統的推薦質量。
침대전통협동과려추천산법재용호평분수거겁단희소정황하무법취득령인만의적추천질량문제,결합User-based 화Item-based 협동과려산법사상,제출료일충기우선택성예측책략적협동과려추천산법,산법이용고상사도역치래계산용호상사성화항목상사성,병통과형성용호최근린거집화항목최근린거집래예측전충평분구진。기우Movielens수거집적실험표명,개진적산법유효개선료전통협동과려추천산법적수거희소성화확전성문제,명현제고료계통적추천질량。
The user rating data in traditional collaborative filtering recommendation algorithm are extremely sparse , which re-sults in poor recommendation quality .A recommendation algorithm based on selective prediction strategy was proposed .The user-based recommendation algorithm was combined with item -based recommendation algorithm .The user similarity and the item similarity were calculated by high similarity threshold and the user -item matrix was evaluated by finding the neighbors of users and items.The experimental results based on MovieLens data set show that the improved algorithm could solve the problem of da -ta sparsity and scalability , and it could improve the accuracy of system recommendation significantly .