基于隐空间映射的带符号网络上的顶点分类
|
摘要
社会网络顶点分类在解决实际问题中有广泛的应用,但绝大多数现有的网络顶点分类算法都集中在无符号的网络,而在边上具有符号的社交网络上的顶点分类算法却很少,且负链接对于符号网络分析的作用大于正链接。研究了符号网络中顶点的分类问题。首先将正、负网络映射到相对应的隐空间,提出基于隐空间的正负链接的数学模型;然后提出优化该模型的迭代算法,通过对隐空间矩阵和映射矩阵的迭代优化,来对网络中的顶点进行分类。由带符号的社会网络数据集的实验结果证明,该算法在数据集Epinions上得到结果的F1值在11以上,在数据集Slashdo上得到结果的F1值在23.8以上,与随机算法相比具有较高的精确度。
Social network node classification is widely used in solving practical problems. Most of the existing network node classification algorithms focus on unsigned social networks,while node classification algorithms on social networks with symbols on edges are rare. Based on the fact that the negative links contribute more on signed network analysis than the positive links. The classification of nodes on signed networks was studied. Firstly, positive and negative networks were projected to the corresponding latent spaces, and a mathematical model was proposed based on positive and negative links in the latent spaces. Then, an iterative algorithm was proposed to optimize the model, and the iterative optimization of latent space matrix and projection matrix was used to classify the nodes in the network. The experimental results on the dataset of the signed social network show that the F1 value of the classification results by the proposed algorithm is higher than 11 on Epinions dataset, and that is higher than 23.8 on Slashdo dataset,which indicate that the proposed algorithm has higher accuracy than random algorithm.
Social network node classification is widely used in solving practical problems. Most of the existing network node classification algorithms focus on unsigned social networks,while node classification algorithms on social networks with symbols on edges are rare. Based on the fact that the negative links contribute more on signed network analysis than the positive links. The classification of nodes on signed networks was studied. Firstly, positive and negative networks were projected to the corresponding latent spaces, and a mathematical model was proposed based on positive and negative links in the latent spaces. Then, an iterative algorithm was proposed to optimize the model, and the iterative optimization of latent space matrix and projection matrix was used to classify the nodes in the network. The experimental results on the dataset of the signed social network show that the F1 value of the classification results by the proposed algorithm is higher than 11 on Epinions dataset, and that is higher than 23.8 on Slashdo dataset,which indicate that the proposed algorithm has higher accuracy than random algorithm.
引文
[1] GUHA R,KUMAR R,RAGHAVAN P,et al.Propagation of trust and distrust[C]// Proceedings of the 13th International Conference on World Wide Web.New York:ACM,2004:403-412.
[2] SONG D J,MEYER D.Link sign prediction and ranking in signed directed social networks[J].Social Network Analysis & Mining,2015,5(1):1-14.
[3] BURKE M,KRAUL R.Mopping up:modeling Wikipedia promotion decisions[C]// Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work.New York:ACM,2008:27-36.
[4] KUNEGIS J,LOMMATZSCH A,BAUCKHAGE C.The Slashdot zoo:mining a social network with negative edges[C]// Proceedings of the 18th International Conference on World Wide Web.New York:ACM,2009:541-550.
[5] 王新月,王兴超,雷雳,等.社交网站在社会认同发展中的作用[J].心理科学进展,2018,26(11):2024-2034.(WANG X Y,WANG X C,LEI L,et al.The role of social networking sites in the development of social identity[J].Advances in Psychological Science,2018,26(11):2024-2034.)
[6] ZHELEVA E,GETOOR L.To join or not to join:the illusion of privacy in social networks with mixed public and private user profiles[C]// Proceedings of the 18th International Conference on World Wide Web.New York:ACM,2009:531-540.
[7] ABBASI M A,TANG J,LIU H.Scalable learning of users preferences using networked data[C]// Proceedings of the 25th ACM Conference on Hypertext and Social Media.New York:ACM,2014:4-12.
[8] TANG L,LIU H.Relational learning via latent social dimensions[C]// Proceedings of the 15th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.New York:ACM,2009:817-826.
[9] SEN P,NAMATA G,BILGIC M,et al.Collective classification in network data[J].AI Magazine,2008,29(3):93.
[10] GETOOR L,DIEHL C P.Link mining:a survey[J].ACM SIGKDD Explorations Newsletter,2005,7(2):3-12.
[11] BHAGAT S,CORMODE G,MUTHUKRISHNAN S.Node classification in social networks[C]// Social Network Data Analytics.Berlin:Springer,2011:115-148.
[12] LU Q,GETOOR L.Link-based classification[C]// Proceedings of the 20th International Conference on International Conference on Machine Learning.Menlo Park,CA:AAAI Press,2003:496-503.
[13] AZRAN A.The rendezvous algorithm:multiclass semi-supervised learning with Markov random walks[C]// Proceedings of the 24th International Conference on Machine Learning.New York:ACM,2007:49-56.
[14] ZHU X,GHAHRAMANI Z,JOHN L,et al.Semi-supervised learning using Gaussian fields and harmonic functions[C]// Proceedings of the 20th International Conference on Machine Learning.Menlo Park,CA:AAAI Press,2003:912-919.
[15] ZHOU D,BOUSQUET O,THOMAS N L,et al.Learning with local and global consistency[C]// Proceedings of the 16th International Conference on Neural Information Processing Systems.Cambridge,MA:MIT Press,2003:321-328.
[16] ZHOU D,HUANG J,SCHOLKOPF B.Learning from labeled and unlabeled data on a directed graph[C]// Proceedings of the 22nd International Conference on Machine Learning.New York:ACM,2005:1036-1043.
[17] BALUJA S,SETH R,SIVAKUMAR D,et al.Video suggestion and discovery for YouTube:taking random walks through the view graph[C]// Proceedings of the 17th International Conference on World Wide Web.New York:ACM,2008:895-904.
[18] NANDANWAR S,MURTY M N.Structural neighborhood based classification of nodes in a network[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM,2016:1085-1094.
[19] TASKAR B,ABBEEL P,KOLLER D.Discriminative probabilistic models for relational data[C]// Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence.New York:ACM,2002:485-492.
[20] 王小攀,胡艳.基于双图结构标签传递算法的高光谱数据分类[J].计算机与数字工程,2018,46(10):2117-2122.(WANG X P,HU Y.Hyperspectral data classification based on double graph structure label transfer algorithm[J].Computer and Digital Engineering,2018,46(10):2117-2122.)
[21] 邵海军.弱关联Web数据分类方法优化研究与仿真[J].计算机仿真,2015,32(12):392-395.(SHAO H J.Research and simulation on the optimization of weak association Web data classification method[J].Computer Simulation,2015,32(12):392-395.)
[22] LI Y C,NIE X Q,HUANG R.Web spam classification method based on deep belief networks[J].Expert Systems with Applications,2018,96:261-270.
[23] ZHANG Z,WANG H,LIU L,et al.Multi-label relational classification via node and label correlation[J].Neurocomputing,2018,292:72-81.
[24] KAZIENKO P,KAJDANOWICZ T.Label-dependent node classification in the network[J].Neurocomputing,2012,75(1):199-209.
[25] ZHANG Q,LI M Z,DENG Y.Measure the structure similarity of nodes in complex networks based on relative entropy[J].Physica A:Statistical Mechanics and its Applications,2018,491:749-763.
[26] GIANVITO P,FRANCESCO S,DONATO M,et al.Multi-type clustering and classification from heterogeneous networks[J].Information Sciences,2018,425:107-126.
[27] LESKOVEC J,HUTTENLOCHER D.Predicting positive and negative links in online social networks[C]// Proceedings of the 19th International Conference on World Wide Web.New York:ACM,2010:641-650.
[2] SONG D J,MEYER D.Link sign prediction and ranking in signed directed social networks[J].Social Network Analysis & Mining,2015,5(1):1-14.
[3] BURKE M,KRAUL R.Mopping up:modeling Wikipedia promotion decisions[C]// Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work.New York:ACM,2008:27-36.
[4] KUNEGIS J,LOMMATZSCH A,BAUCKHAGE C.The Slashdot zoo:mining a social network with negative edges[C]// Proceedings of the 18th International Conference on World Wide Web.New York:ACM,2009:541-550.
[5] 王新月,王兴超,雷雳,等.社交网站在社会认同发展中的作用[J].心理科学进展,2018,26(11):2024-2034.(WANG X Y,WANG X C,LEI L,et al.The role of social networking sites in the development of social identity[J].Advances in Psychological Science,2018,26(11):2024-2034.)
[6] ZHELEVA E,GETOOR L.To join or not to join:the illusion of privacy in social networks with mixed public and private user profiles[C]// Proceedings of the 18th International Conference on World Wide Web.New York:ACM,2009:531-540.
[7] ABBASI M A,TANG J,LIU H.Scalable learning of users preferences using networked data[C]// Proceedings of the 25th ACM Conference on Hypertext and Social Media.New York:ACM,2014:4-12.
[8] TANG L,LIU H.Relational learning via latent social dimensions[C]// Proceedings of the 15th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.New York:ACM,2009:817-826.
[9] SEN P,NAMATA G,BILGIC M,et al.Collective classification in network data[J].AI Magazine,2008,29(3):93.
[10] GETOOR L,DIEHL C P.Link mining:a survey[J].ACM SIGKDD Explorations Newsletter,2005,7(2):3-12.
[11] BHAGAT S,CORMODE G,MUTHUKRISHNAN S.Node classification in social networks[C]// Social Network Data Analytics.Berlin:Springer,2011:115-148.
[12] LU Q,GETOOR L.Link-based classification[C]// Proceedings of the 20th International Conference on International Conference on Machine Learning.Menlo Park,CA:AAAI Press,2003:496-503.
[13] AZRAN A.The rendezvous algorithm:multiclass semi-supervised learning with Markov random walks[C]// Proceedings of the 24th International Conference on Machine Learning.New York:ACM,2007:49-56.
[14] ZHU X,GHAHRAMANI Z,JOHN L,et al.Semi-supervised learning using Gaussian fields and harmonic functions[C]// Proceedings of the 20th International Conference on Machine Learning.Menlo Park,CA:AAAI Press,2003:912-919.
[15] ZHOU D,BOUSQUET O,THOMAS N L,et al.Learning with local and global consistency[C]// Proceedings of the 16th International Conference on Neural Information Processing Systems.Cambridge,MA:MIT Press,2003:321-328.
[16] ZHOU D,HUANG J,SCHOLKOPF B.Learning from labeled and unlabeled data on a directed graph[C]// Proceedings of the 22nd International Conference on Machine Learning.New York:ACM,2005:1036-1043.
[17] BALUJA S,SETH R,SIVAKUMAR D,et al.Video suggestion and discovery for YouTube:taking random walks through the view graph[C]// Proceedings of the 17th International Conference on World Wide Web.New York:ACM,2008:895-904.
[18] NANDANWAR S,MURTY M N.Structural neighborhood based classification of nodes in a network[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM,2016:1085-1094.
[19] TASKAR B,ABBEEL P,KOLLER D.Discriminative probabilistic models for relational data[C]// Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence.New York:ACM,2002:485-492.
[20] 王小攀,胡艳.基于双图结构标签传递算法的高光谱数据分类[J].计算机与数字工程,2018,46(10):2117-2122.(WANG X P,HU Y.Hyperspectral data classification based on double graph structure label transfer algorithm[J].Computer and Digital Engineering,2018,46(10):2117-2122.)
[21] 邵海军.弱关联Web数据分类方法优化研究与仿真[J].计算机仿真,2015,32(12):392-395.(SHAO H J.Research and simulation on the optimization of weak association Web data classification method[J].Computer Simulation,2015,32(12):392-395.)
[22] LI Y C,NIE X Q,HUANG R.Web spam classification method based on deep belief networks[J].Expert Systems with Applications,2018,96:261-270.
[23] ZHANG Z,WANG H,LIU L,et al.Multi-label relational classification via node and label correlation[J].Neurocomputing,2018,292:72-81.
[24] KAZIENKO P,KAJDANOWICZ T.Label-dependent node classification in the network[J].Neurocomputing,2012,75(1):199-209.
[25] ZHANG Q,LI M Z,DENG Y.Measure the structure similarity of nodes in complex networks based on relative entropy[J].Physica A:Statistical Mechanics and its Applications,2018,491:749-763.
[26] GIANVITO P,FRANCESCO S,DONATO M,et al.Multi-type clustering and classification from heterogeneous networks[J].Information Sciences,2018,425:107-126.
[27] LESKOVEC J,HUTTENLOCHER D.Predicting positive and negative links in online social networks[C]// Proceedings of the 19th International Conference on World Wide Web.New York:ACM,2010:641-650.