| 190 | 0 | 266 |
| 下载次数 | 被引频次 | 阅读次数 |
在经典网络表示学习框架上进行改进，提出了基于文本注意力机制优化的网络表示学习模型。首先学习上下文节点的平均嵌入，然后利用上下文节点的平均嵌入引入注意力机制，目标节点的嵌入由注意力和文本嵌入共同决定。在文本特征上添加注意力机制，旨在为文本特征中的词语学习不同的权重值，使得对模型有利的词语得到最大贡献，有效避免低频词、噪声词对模型的影响。在Citeseer (M10)、DBLP (V4)和SDBLP三个数据集上进行实验，结果表明，该模型的网络节点分类性能优于DeepWalk算法和同类别表示学习算法。在网络可视化分析中，所提算法有明显的聚类现象和聚类边界，获得了期望的结果。
Abstract：To improve the classical network representation learning framework, a network representation learning model based on text attention mechanism optimization was proposed.The average embedding of context nodes was first learned. Then the average embedding of context nodes was used to introduce the attention mechanism. And the embedding of target nodes was determined by the attention and text embedding together. Attention mechanism was added to the text features to learn different weight values for words in the text features, so that the words that were beneficial to the model could get the maximum contribution and effectively avoid the influence of low-frequency words and noisy words on the model. Experiments on Citeseer(M10), DBLP(V4), and SDBLP datasets showed that the network node classification performance of this model was superior to DeepWalk algorithm and similar representation learning algorithms. In the network visualization analysis, the proposed algorithm had obvious clustering phenomenon and clustering boundary, and obtained the desired results.
[1] 涂存超，杨成，刘知远，等.网络表示学习综述[J].中国科学(信息科学),2017,47(8):980-996.TU C C,YANG C,LIU Z Y,et al.Network representation learning:an overview[J].Scientia sinica (informationis),2017,47(8):980-996.
[2] PEROZZI B,AL-RFOU R,SKIENA S.DeepWalk:online learning of social representations[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM Press,2014:701-710.
[3] GROVER A,LESKOVEC J.Node2vec:scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM Press,2016:855-864.
[4] TANG J,QU M,WANG M Z,et al.LINE:large-scale information network embedding[C]//Proceedings of the 24th International Conference on World Wide Web.Berlin:Springer Press,2015:1067-1077.
[5] RIBEIRO L F R,SAVERESE P H P,FIGUEIREDO D R.Struc2vec:learning node representations from structural identity[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM Press,2017:385-394.
[6] TU C C,ZHANG Z Y,LIU Z Y,et al.TransNet:translation-based network representation learning for social relation extraction[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence.Palo Alto:AAAI Press,2017:2864-2870.
[7] LI Z C,TANG J H,ZHANG L Y,et al.Weakly-supervised semantic guided hashing for social image retrieval[J].International journal of computer vision,2020,128(8/9):2265-2278.
[8] PAN S,WU J,ZHU X,et al.Tri-party deep network representation[C]//Proceedings of the 25th International Joint Conference on Artifical Intelligence.San Francisco:Morgan Kaufmann Publishers,2016:1895-1901.
[9] 冶忠林，赵海兴，张科，等.基于邻节点和关系模型优化的网络表示学习[J].计算机研究与发展，2019,56(12):2562-2577.YE Z L,ZHAO H X,ZHANG K,et al.Network representation learning using the optimizations of neighboring vertices and relation model[J].Journal of computer research and development,2019,56(12):2562-2577.
[10] 窦伟，张维玉，翁自强，等.结构增强的属性网络表示学习[J].计算机工程与应用，2021,57(18):149-155.DOU W,ZHANG W Y,WENG Z Q,et al.Structure-enhanced attributed network representation learning[J].Computer engineering and applications,2021,57(18):149-155.
[11] 倪琦瑄，张霞，卜湛.基于双视角的耦合网络表示学习算法[J].计算机系统应用，2021,30(9):247-255.NI Q X,ZHANG X,BU Z.Coupled network embedding method based on dual perspectives[J].Computer systems & applications,2021,30(9):247-255.
[12] 陈斌，李金龙.基于注意力机制融合多阶邻域信息的网络表示学习模型[J].小型微型计算机系统，2021,42(4):761-765.CHEN B,LI J L.Attention-based network representation learning model using multi-neighboring information[J].Journal of Chinese computer systems,2021,42(4):761-765.
[13] 樊玮，王慧敏，邢艳.基于自编码器的多视图属性网络表示学习模型[J].计算机应用，2021,41(4):1064-1070.FAN W,WANG H M,XING Y.Auto-encoder based multi-view attributed network representation learning model[J].Journal of computer applications,2021,41(4):1064-1070.
[14] WANG W,MA D Y,XIN G D,et al.A network representation learning method based on topology[J].Information sciences,2021,571:443-458.
[15] 张潘，卢光跃，吕少卿，等.基于矩阵分解的属性网络表示学习[J].计算机工程，2020,46(10):67-73.ZHANG P,LU G Y,LYU S Q,et al.Attributed network representation learning based on matrix factorization[J].Computer engineering,2020,46(10):67-73.
[16] 康世泽，吉立新，张建朋.一种基于图注意力网络的异质信息网络表示学习框架[J].电子与信息学报，2021,43(4):915-922.KANG S Z,JI L X,ZHANG J P.Heterogeneous information network representation learning framework based on graph attention network[J].Journal of electronics & information technology,2021,43(4):915-922.
[17] MIKOLOV T,SUTSKEVER I,CHEN K,et al.Distributed representations of words and phrases and their compositionality[C]//Proceedings of the 27th Annual Conference on Neural Information Processing Systems.Cambridge:MIT Press,2013:3111-3119.
[18] MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[EB/OL].(2013-01-16)[2022-01-15].https://arxiv.org/pdf/1301.3781.
[19] MIKOLOV T,KARAFIáT M,BURGET L,et al.Recurrent neural network based language model[C]//Proceedings of International Speech Communication Assciation.Amsterdam:Elsevier Press,2010:1045-1048.
[20] YANG C,LIU Z,ZHAO D,et al.Network representation learning with rich text information[C]//Proceedings of the 24th International Joint Conference on Artifical Intelligence.Amsterdam:Elsevier Press,2015:2111-2117.
[21] FAN R E,CHANG K W,HSIEH C J,et al.LIBLINEAR:a library for large linear classification[J].Journal of machine learning research,2008,9:1871-1874.
基本信息:
DOI：10.13705/j.issn.1671-6841.2022156
中图分类号:TP391.1
引用信息:
[1]唐彦龙,冶忠林,赵海兴,等.基于文本注意力机制优化的网络表示学习模型[J],2023,55(06):41-47.DOI:10.13705/j.issn.1671-6841.2022156.
基金信息:
国家重点研发计划项目(2020YFC1523300);; 国家自然科学基金青年科学基金项目(62007019);; 青海省自然科学基金青年项目(2021-ZJ-946Q)