
软件学报 ISSN 1000-9825, CODEN RUXUEW E-mail: jos@iscas.ac.cn
Journal of Software, 2022,33(2):498523 [doi: 10.13328/j.cnki.jos.006353] http://www.jos.org.cn
©中国科学院软件研究所版权所有. Tel: +86-10-62562563
超图学习综述: 算法分类与应用分析
胡秉德
,
王新根
,
王新宇
,
宋明黎
,
陈
纯
(浙江大学 计算机科学与技术学院, 浙江 杭州 310007)
通信作者: 王新根, E-mail: newroot@zju.edu.cn
摘 要: 随着图结构化数据挖掘的兴起, 超图作为一种特殊的图结构化数据, 在社交网络分析、图像处理、生物
反应解析等领域受到广泛关注. 研究者通过解析超图中的拓扑结构与节点属性等信息, 能够有效解决实际应用场
景中所遇到的如兴趣推荐、社群划分等问题. 根据超图学习算法的设计特点, 将其划分为谱分析方法和神经网络
方法, 根据方法对超图处理的不同手段, 可进一步划分为展开式方法和非展开式方法. 若将展开式方法用于不可
分解超图, 则很有可能会造成信息损失. 然而, 现有的超图相关综述文章鲜有就超图学习方法适用于哪类超图这
一问题做出相关归纳. 因此, 分别从超图上的谱分析方法和神经网络方法两方面出发, 对展开式方法和非展开式
方法展开讨论, 并结合其算法特性和应用场景作进一步细分; 然后, 分析比较各类算法的设计思路, 结合实验结
果总结各类算法的优缺点; 最后, 对超图学习未来可能的研究方向进行了展望.
关键词: 超图学习; 谱分析; 神经网络; 展开; 非展开
中图法分类号: TP181
中文引用格式: 胡秉德, 王新根, 王新宇, 宋明黎, 陈纯. 超图学习综述: 算法分类与应用分析. 软件学报, 2022,33(2):
498–523. http://www.jos.org.cn/1000-9825/6353.htm
英文引用格式: Hu BD, Wang XG, Wang XY, Song ML, Chen C. Survey on Hypergraph Learning: Algorithm Classification and
Application Analysis. Ruan Jian Xue Bao/Journal of Software, 2022, 33(2): 498523 (in Chinese). http://www.jos.org.cn/1000-9825/
6353.htm
Survey on Hypergraph Learni ng: Algorithm Classific ation and Application Analysis
HU Bing-De, WANG Xin-Gen, WANG Xin-Yu, SONG Ming-Li, CHEN Chun
(College of Computer Science and Technology, Zhejiang University, Hangzhou 310007, China)
Abstra ct : With the rise of graph structured data mining, hypergraph, as a special type of graph structured data, is widely concerned in
social network analysis, image processing, biological response analysis, and other fields. By analyzing the topological structure and node
attributes of hypergraph, many problemscan be effectively solved such as recommendation, community detection, and so on. According to
the characteristics of hypergraph learning algorithm, it can be divided into spectral analysis method, neural network method, and other
method. According to the methods used to process hypergraphs, it can be further divided into expansion method and non-expansion
method. If the expansion method is applied to the indecomposable hypergraph, it is likely to cause information loss. However, the existing
hypergraph reviews do not discuss that hypergraph learning methods are applicable to which type of hypergraphs. So, this article discusses
the expansion method and non-expansion method respectively from the aspects of spectral analysis method and neural network method,
and further subdivides them according to their algorithm characteristics and application scenarios. Then, the ideas of different algorithms
are analyzed and comparedin experiments. The advantages and disadvantages of different algorithms are concluded. Finally, some
promising research directionsare proposed.
Key words: hypergraph learning; spectral analysis; neural network; expansion; non-expansion
图(graph)作为一种高效的关系表达结构, 被广泛地应用于成对关系的建模中, 例如对论文引用关系、私人
基金项目: 广东省重点领域研发计划(2020B0101100005); 浙江省重点研发计划(2021C01014)
收稿时间: 2020-08-07; 修改时间: 2020-09-30; 采用时间: 2021-04-17; jos 在线出版时间: 2021-05-20
评论