基于多核学习的协同滤波算法
|
摘要
协同滤波是当前推荐系统中一种主流的个性化推荐算法,通过近似用户对商品的评价进行推荐。核函数是解决非线性模式问题的一种方法。协同滤波通常会选用不同的核函数来分析用户之间的影响关系。由于单核函数无法适应于复杂多变场景。因此,结合多个核函数成为一种解决方法。多核学习能够针对场景来组合各个核函数以获取更好的结果。本文提出了一种基于多核学习的协同滤波算法。该算法在现有核函数的基础上,优化各个核函数的权重以匹配数据的分布。在大众点评数据集和Foursquare数据集上的实验结果表明:基于多核学习的协同滤波算法比经验给定的相似函数的性能要高,具有更好的普适性。
As a frequently personalized recommendation algorithm of the currently recommendation system,collaborative filtering uses the item evaluation by the approximate users to recommend. Kernel function is an approach for non-linear pattern analysis problems. Ordinarily,collaborative filtering will choose some different kernel functions to analyse the influence between the users. Since the single kernel function can not be adapted to the complicated and various scene,the combination of multiply kernel function becomes a solution. In terms of scenes,multiply kernel learning can combine every kernel function for a better result. This paper proposes a collaborative filtering algorithm based on multiple kernel learning. Based on the available kernel function,this algorithm optimizes the weights of every kernel function to match the data distribution. The experimental result on dianping dataset and foursquare dataset shows that compared with the collaborative filtering algorithm based on common similarity,the collaborative filtering algorithm based on multiple kernel learning achieves better performance. That is,multiple kernel learning has a better common adaptation.
As a frequently personalized recommendation algorithm of the currently recommendation system,collaborative filtering uses the item evaluation by the approximate users to recommend. Kernel function is an approach for non-linear pattern analysis problems. Ordinarily,collaborative filtering will choose some different kernel functions to analyse the influence between the users. Since the single kernel function can not be adapted to the complicated and various scene,the combination of multiply kernel function becomes a solution. In terms of scenes,multiply kernel learning can combine every kernel function for a better result. This paper proposes a collaborative filtering algorithm based on multiple kernel learning. Based on the available kernel function,this algorithm optimizes the weights of every kernel function to match the data distribution. The experimental result on dianping dataset and foursquare dataset shows that compared with the collaborative filtering algorithm based on common similarity,the collaborative filtering algorithm based on multiple kernel learning achieves better performance. That is,multiple kernel learning has a better common adaptation.
引文
[1]Badrul Sarwar.Item-based collaborative filtering recommendation algorithms[C]∥Proceedings of the 10thInternational Conference on World Wide Web.New York:WWW10,2001:285-295.
[2]Suykens J K.Least squares support vector machine classifiers[J].Neural Processing Letters,1993,9(3):293-300.
[3]汪洪桥.多核学习方法[J].自动化学报,2010,36(8):1037-1050.Wang Hongqiao.On multiple kernel learning methods[J].Acta Automatic Sinica,2010,36(8):1037-1050.
[4]Meirom E A.Nuc-MKL:A convex approach to non linear multiple kernel learning[J].AISTATS,2016,51:610-619.
[5]Li Liyan.Gene networks identification using independence measurement based on the Hilbert space[D].Hangzhou:Hangzhou Dianzi University,2014.
[6]Lyu Siwei.Mercer kernels for object recognition with local features[J].CVPR,2005,2:1063-6919.
[7]Liu Haifeng.A new user similarity model toimprove the accuracy of collaborative filtring[J].Knowledge-Based Systems,2014,56:156-166.
[8]Ekstrand M D,Riedl J T,Konstan J A.Collaborative filtering recommender system[J].Foundations and Trends in Human-Computer Interaction,2010,4(2):81-173.
[9]Chung Kai-Min.Radius margin bounds for support vector machines with the RBF kernel[J].Neural Computation,2003,15(11):2643-2681.
[10]奚吉,赵力,左加阔.基于改进多核学习的语音情感识别算法[J].数据采集与处理,2014,29(5):730-734.Xi Ji,Zhao Li,Zuo Jiahuo.Speech emotion recognition based on modified multiple kernel learning algorithm[J].Journal of Data Acquisition and Processing,2014,29(5):730-734.
[11]王国胜.核函数的性质及其构造方法[J].计算机科学,2006,33(6):172-178.Wang Guosheng.Properties and construction methods of kernel in support vector machine[J].Computer Science,2006,33(6):172-178.
[12]Lu Yanting.Research and application of clustering and hierarchical classification algorithms based on multiple kernel learning[D].Nanjing:Nanjing University of Science and Technology,2013.
[13]王付强,基于位置的非对称相似性度量的协同过滤推荐算法[J].计算机应用,2016,36(1):171-174.Wang Fuqiang.Location-based asymmetric similarity for collaborative filtering recommendation algorithm[J].Journal of Computer Applications,2016,36(1):171-174.
[14]Léon Bottou.Large-scale machine learning with stochastic gradient descent[C]∥19th International Conference on Computational Statistics.Paris:Proceedings of COMPSTAT,2010:177-186.
[15]York D.Least squares fitting of a straight line with correlated errors[J].Earth and Planetary Science Letters,1968,5:320-324.
[16]Yang Yiming.A re-examination of text categorization methods[C]∥the 22nd Annual International ACM SIGIR Conference.USA:SIGIR,1999:42-49.
[17]Vassiliadis S.The sum-absolute-difference motion estimation accelerator[C]∥Proceedings of the 24thEuromicro Conference.Germany:Euromicro Conference,1998:559-566.
[2]Suykens J K.Least squares support vector machine classifiers[J].Neural Processing Letters,1993,9(3):293-300.
[3]汪洪桥.多核学习方法[J].自动化学报,2010,36(8):1037-1050.Wang Hongqiao.On multiple kernel learning methods[J].Acta Automatic Sinica,2010,36(8):1037-1050.
[4]Meirom E A.Nuc-MKL:A convex approach to non linear multiple kernel learning[J].AISTATS,2016,51:610-619.
[5]Li Liyan.Gene networks identification using independence measurement based on the Hilbert space[D].Hangzhou:Hangzhou Dianzi University,2014.
[6]Lyu Siwei.Mercer kernels for object recognition with local features[J].CVPR,2005,2:1063-6919.
[7]Liu Haifeng.A new user similarity model toimprove the accuracy of collaborative filtring[J].Knowledge-Based Systems,2014,56:156-166.
[8]Ekstrand M D,Riedl J T,Konstan J A.Collaborative filtering recommender system[J].Foundations and Trends in Human-Computer Interaction,2010,4(2):81-173.
[9]Chung Kai-Min.Radius margin bounds for support vector machines with the RBF kernel[J].Neural Computation,2003,15(11):2643-2681.
[10]奚吉,赵力,左加阔.基于改进多核学习的语音情感识别算法[J].数据采集与处理,2014,29(5):730-734.Xi Ji,Zhao Li,Zuo Jiahuo.Speech emotion recognition based on modified multiple kernel learning algorithm[J].Journal of Data Acquisition and Processing,2014,29(5):730-734.
[11]王国胜.核函数的性质及其构造方法[J].计算机科学,2006,33(6):172-178.Wang Guosheng.Properties and construction methods of kernel in support vector machine[J].Computer Science,2006,33(6):172-178.
[12]Lu Yanting.Research and application of clustering and hierarchical classification algorithms based on multiple kernel learning[D].Nanjing:Nanjing University of Science and Technology,2013.
[13]王付强,基于位置的非对称相似性度量的协同过滤推荐算法[J].计算机应用,2016,36(1):171-174.Wang Fuqiang.Location-based asymmetric similarity for collaborative filtering recommendation algorithm[J].Journal of Computer Applications,2016,36(1):171-174.
[14]Léon Bottou.Large-scale machine learning with stochastic gradient descent[C]∥19th International Conference on Computational Statistics.Paris:Proceedings of COMPSTAT,2010:177-186.
[15]York D.Least squares fitting of a straight line with correlated errors[J].Earth and Planetary Science Letters,1968,5:320-324.
[16]Yang Yiming.A re-examination of text categorization methods[C]∥the 22nd Annual International ACM SIGIR Conference.USA:SIGIR,1999:42-49.
[17]Vassiliadis S.The sum-absolute-difference motion estimation accelerator[C]∥Proceedings of the 24thEuromicro Conference.Germany:Euromicro Conference,1998:559-566.