一种基于稀疏分段的协同过滤推荐算法
|
摘要
针对数据强稀疏性严重制约协同过滤算法推荐准确性的问题,提出基于稀疏分段的改进方法。首先利用基于迭代预测的支持向量回归在解决小样本高维数据中的优势,对稀疏的U-I矩阵中相对弱稀疏的密集数据部分预测缺失评分,然后使用基于项目的插补协同过滤方法预测剩余数据的缺失评分。在多个公开数据集中的实验表明,该方法适用于强稀疏数据集的推荐,与基于项目协同过滤比较可取得较好的预测结果。
Since the strong sparsity of data seriously affects the recommendation accuracy of collaborative filtering algorithm,an improved method based on sparseness segmentation is put forward. The support vector regression based on iterative prediction is used to estimate the missing scores for the relatively-weak sparse density data of sparse U-I matrix,which has the advantage in predicting the high-dimensional small-sample data. The item-based imputative collaborative filtering method is employed to predict the missing scores of residue data. The experimental results of several public datasets show that the method is suitable for the recommendation of strong-sparse dataset,and can obtain better prediction results than the item-based collaborative filtering method.
Since the strong sparsity of data seriously affects the recommendation accuracy of collaborative filtering algorithm,an improved method based on sparseness segmentation is put forward. The support vector regression based on iterative prediction is used to estimate the missing scores for the relatively-weak sparse density data of sparse U-I matrix,which has the advantage in predicting the high-dimensional small-sample data. The item-based imputative collaborative filtering method is employed to predict the missing scores of residue data. The experimental results of several public datasets show that the method is suitable for the recommendation of strong-sparse dataset,and can obtain better prediction results than the item-based collaborative filtering method.
引文
[1]EKSTRAND M D,RIEDL J T,KONSTAN J A.Collaborative filtering recommender systems[J].Foundations and trends in human-computer interaction,2011,4(2):81-173.
[2]KOREN Y.Factorization meets the neighborhood:a multifaceted collaborative filtering model[C]//Proceedings of the 14th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas:ACM,2008:426-434.
[3]KOREN Y,BELL R.Advances in collaborative filtering[M]//Anon.Recommender systems handbook.US:Springer,2015:77-118.
[4]SU X,KHOSHGOFTAAR T M,GREINER R.A mixture imputation-boosted collaborative filter[C]//Proceedings of the 21st International Florida Artificial Intelligence Research Society Conference.Florida:ACM,2008:312-317.
[5]REN Y,LI G,ZHANG J,et al.The efficient imputation method for neighborhood-based collaborative filtering[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.[S.l.]:ACM,2012:684-693.
[6]SHI Y,LARSON M,HANJALIC A.Collaborative filtering beyond the user-item matrix:a survey of the state of the art and future challenges[J].ACM computing surveys,2014,47(1):3.
[7]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.DING Shifei,QI Bingjuan,TAN Hongyan.An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China,2011,40(1):2-10.
[8]CHRISTAKOPOULOU E,KARYPIS G.Local item-item models for top-n recommendation[C]//Proceedings of the 10th ACMConference on Recommender Systems.[S.l.]:ACM,2016:67-74.
[9]MovieLens.GroupLens datasheet[EB/OL].[2015-12-15].https://grouplens.org/datasets/movielens/.
[10]EachMovie.EachMovie collaborative filtering data set[EB/OL].[1997-02-16].http://www.cs.cmu.edu/~lebanon/IR-lab/data.html.
[11]LI Z,PENG J Y,GENG G H,et al.Video recommendation based on multi-modal information and multiple kernel[J].Multimedia tools and applications,2015,74(13):4599-4616.
[12]CHANG C C,LIN C J.LIBSVM:a library for support vector machines[J].ACM transactions on intelligent systems and technology,2011,2(3):112-115.
[13]WU X,CHENG B,CHEN J.Collaborative filtering service recommendation based on a novel similarity computation method[J].IEEE transactions on services computing,2017,10(3):352-365.
[14]HU Y,SHI W,LI H,et al.Mitigating data sparsity using similarity reinforcement-enhanced collaborative filtering[J].ACM transactions on Internet technology,2017,17(3):1-20.
[2]KOREN Y.Factorization meets the neighborhood:a multifaceted collaborative filtering model[C]//Proceedings of the 14th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas:ACM,2008:426-434.
[3]KOREN Y,BELL R.Advances in collaborative filtering[M]//Anon.Recommender systems handbook.US:Springer,2015:77-118.
[4]SU X,KHOSHGOFTAAR T M,GREINER R.A mixture imputation-boosted collaborative filter[C]//Proceedings of the 21st International Florida Artificial Intelligence Research Society Conference.Florida:ACM,2008:312-317.
[5]REN Y,LI G,ZHANG J,et al.The efficient imputation method for neighborhood-based collaborative filtering[C]//Proceedings of the 21st ACM international conference on Information and knowledge management.[S.l.]:ACM,2012:684-693.
[6]SHI Y,LARSON M,HANJALIC A.Collaborative filtering beyond the user-item matrix:a survey of the state of the art and future challenges[J].ACM computing surveys,2014,47(1):3.
[7]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.DING Shifei,QI Bingjuan,TAN Hongyan.An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China,2011,40(1):2-10.
[8]CHRISTAKOPOULOU E,KARYPIS G.Local item-item models for top-n recommendation[C]//Proceedings of the 10th ACMConference on Recommender Systems.[S.l.]:ACM,2016:67-74.
[9]MovieLens.GroupLens datasheet[EB/OL].[2015-12-15].https://grouplens.org/datasets/movielens/.
[10]EachMovie.EachMovie collaborative filtering data set[EB/OL].[1997-02-16].http://www.cs.cmu.edu/~lebanon/IR-lab/data.html.
[11]LI Z,PENG J Y,GENG G H,et al.Video recommendation based on multi-modal information and multiple kernel[J].Multimedia tools and applications,2015,74(13):4599-4616.
[12]CHANG C C,LIN C J.LIBSVM:a library for support vector machines[J].ACM transactions on intelligent systems and technology,2011,2(3):112-115.
[13]WU X,CHENG B,CHEN J.Collaborative filtering service recommendation based on a novel similarity computation method[J].IEEE transactions on services computing,2017,10(3):352-365.
[14]HU Y,SHI W,LI H,et al.Mitigating data sparsity using similarity reinforcement-enhanced collaborative filtering[J].ACM transactions on Internet technology,2017,17(3):1-20.