Hypergraph-based eaves tile dating method under data imbalance conditions

doi:10.11772/j.issn.1001-9081.2025010030

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (2): 620-629.DOI: 10.11772/j.issn.1001-9081.2025010030

• Frontier and comprehensive applications • Previous Articles

Hypergraph-based eaves tile dating method under data imbalance conditions

Xing QIU¹, Zuxing XUAN²(), Kejia HUANG³, Wen ZHANG³, Xiao ZHUANG²

^1.Smart City College，Beijing Union University，Beijing 100101，China
^2.Institute of Fundamental and Interdisciplinary Sciences，Beijing Union University，Beijing 100101，China
^3.Institute of Archaeology，Beijing Union University，Beijing 100101，China

Received:2025-01-10 Revised:2025-04-18 Accepted:2025-04-18 Online:2025-05-26 Published:2026-02-10
Contact: Zuxing XUAN
About author:QIU Xing， born in 1998， M. S. candidate. His research interests include graph deep learning， hypergraph learning.
XUAN Zuxing， born in 1980， Ph. D.， professor. His research interests include hypergraph learning， applied mathematics. Email:zuxingxuan@163.com
HUANG Kejia， born in 1978， M. S.， professor. His research interests include Chinese archaeology， cultural heritage.
ZHANG Wen， born in 1982， Ph. D.， associate professor. Her research interests include Chinese archaeology， cultural heritage.
ZHUANG Xiao， born in 1990， Ph. D.， lecturer. His research interests include algebraic representation theory， hypergraph learning.
Supported by:
Key Research Funding Program of Beijing Union University(ZKZD202306)

基于超图的数据不平衡条件下的瓦当年代判别方法

邱星¹, 玄祖兴²(), 黄可佳³, 张雯³, 庄晓²

^1.北京联合大学智慧城市学院，北京 100101
^2.北京联合大学数理与交叉科学研究院，北京 100101
^3.北京联合大学考古研究院，北京 100101

通讯作者: 玄祖兴
作者简介:邱星（1998—），男，河南唐河人，硕士研究生，主要研究方向：图深度学习、超图学习
玄祖兴（1980—），男，山东潍坊人，教授，博士，主要研究方向：超图学习、应用数学Email:zuxingxuan@163.com
黄可佳（1978—），男，河南新野人，教授，硕士，主要研究方向：中国考古学、文化遗产
张雯（1982—），女，山东济南人，副教授，博士，主要研究方向：中国考古学、文化遗产
庄晓（1990—），男，河南南阳人，讲师，博士，主要研究方向：代数表示论、超图学习。
基金资助:
北京联合大学重点科研资助项目(ZKZD202306)

Abstract

Abstract:

To address inefficiency and subjectivity of manual eaves tile dating methods， a Hypergraph-based Eaves Tile Dating method under Data Imbalance Conditions （HETD-DIC） was proposed to provide a more objective assistive tool for archaeological dating. Firstly， a dual-weight computation mechanism was designed， which means that a hyperedge weight computation module was used to aggregate associated node features， so as to generate hyperedge weights， and these hyperedge weights were then used to calculate node weights， which reduced the impact of imbalanced sample distribution. Secondly， a Hyperedge Node Relation Matrix （HNRM） was constructed to establish a feature encoding-decoding channel， so as to enhance the representation capability of nodes. Finally， extensive experiments were conducted to evaluate different models， and UniGIN （Unified Graph Isomorphism Network） was selected as baseline classification model due to its superior performance. Experimental results demonstrate that on the self-built eaves tile dataset， with only 20% of training data used， HETD-DIC achieves improvements of 4.67， 4.55， 4.67， and 5.09 percentage points， respectively， compared to UniGIN in accuracy， weighted precision， weighted recall， and weighted F1-score. It can be seen that HETD-DIC solves data imbalance problem effectively and provides reliable automatic assisted decision basis for archaeological dating.

Key words: eaves tile dating, hypergraph, data imbalance, node representation, archaeological dating

摘要：

针对人工瓦当年代判别方法效率低且主观性强的问题，提出一种基于超图的数据不平衡条件下的瓦当年代判别方法（HETD-DIC），以提供更客观的考古辅助判别工具。首先，设计双权重计算机制，利用超边权重计算模块聚合关联节点特征生成超边权重，再利用超边权重生成节点权重，降低样本分布不均产生的影响；其次，构建超边节点关系矩阵（HNRM），建立节点特征编码-解码通道，增强节点的表征能力；最后，通过大量实验评估不同模型，并选取表现较好的UniGIN （Unified Graph Isomorphism Network）作为基础分类模型。实验结果表明，在自建的瓦当数据集上，当仅使用20%的训练数据时， HETD-DIC的准确率、加权精确率、加权召回率和加权F1-score相较于UniGIN分别提升了4.67、4.55、4.67和5.09个百分点。HETD-DIC能有效解决数据不平衡问题，为考古断代提供可靠的自动化辅助决策依据。

关键词: 瓦当年代判别, 超图, 数据不平衡, 节点表征, 考古断代

CLC Number:

TP391.4

Xing QIU, Zuxing XUAN, Kejia HUANG, Wen ZHANG, Xiao ZHUANG. Hypergraph-based eaves tile dating method under data imbalance conditions[J]. Journal of Computer Applications, 2026, 46(2): 620-629.

邱星, 玄祖兴, 黄可佳, 张雯, 庄晓. 基于超图的数据不平衡条件下的瓦当年代判别方法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 620-629.

Figures/Tables 14

References 38

[1]	惠姣姣，曹红宝.基于机器学习分类算法的玻璃文物鉴定［J］.现代信息科技， 2023， 7（13）： 101-104.
	HUI J J， CAO H B. Identification of glass cultural relics based on machine learning classification algorithm ［J］. Modern Information Technology， 2023， 7（13）： 101-104.
[2]	KOONCE B. ResNet 50 ［M］. Convolutional neural networks with swift for tensorflow： image recognition and dataset categorization. Berkeley： Apress， 2021： 63-75.
[3]	孙选铭，苏淼.基于深度学习的丝绸文物纹样识别应用［J］.丝绸， 2023， 60（8）： 1-10.
	SUN X M， SU M. Application of silk cultural relic pattern recognition based on deep learning ［J］. Journal of Silk， 2023， 60（8）： 1-10.
[4]	王鑫，程远，张若愚，等.基于高光谱成像技术的石质文物风化赋存环境分类方法［J］.激光与光电子学进展， 2025， 62（10）： No.1037005.
	WANG X， CHENG Y， ZHANG R Y， et al. Classification of weathering environments for stone cultural heritage based on hyperspectral imaging technology ［J］. Laser and Optoelectronics Progress， 2025， 62（10）： No.1037005.
[5]	WEI M， CHEN G， DING X， et al. Research on classification method of broken pottery in Gansu Province by GNN ［C］// Proceedings of the IEEE 6th International Conference on Information Systems and Computer Aided Education. Piscataway： IEEE， 2023： 757-760.
[6]	AHMED Z， DAS S. A comparative analysis on recent methods for addressing imbalance classification ［J］. SN Computer Science， 2023， 5（1）： No.30.
[7]	ZHAO J， JIN J， CHEN S， et al. A weighted hybrid ensemble method for classifying imbalanced data ［J］. Knowledge-Based Systems， 2020， 203： No.106087.
[8]	ZHANG J， WANG T， NG W W Y， et al. Ensembling perturbation-based oversamplers for imbalanced datasets ［J］. Neurocomputing， 2022， 479： 1-11.
[9]	LIU J. Fuzzy support vector machine for imbalanced data with borderline noise ［J］. Fuzzy Sets and Systems， 2021， 413： 64-73.
[10]	TAO X， LI Q， REN C， et al. Affinity and class probability-based fuzzy support vector machine for imbalanced data sets ［J］. Neural Networks， 2020， 122： 289-307.
[11]	SALEHI A R， KHEDMATI M. A Cluster-based SMOTE Both-sampling （CSBBoost） ensemble algorithm for classifying imbalanced data ［J］. Scientific Reports， 2024， 14： No.5152.
[12]	SINGH H， KAUR M， SINGH B. A hybrid feature weighting and selection-based strategy to classify the high-dimensional and imbalanced medical data ［J］. Neural Computing and Applications， 2024， 36（20）： 12299-12316.
[13]	华玉冰，杨荣昌.姜女石：秦行宫遗址发掘报告（上册）［M］.北京：文物出版社， 2010： 261-266.
	HUA Y B， YANG R C. Excavation report of the Jiangnüshi site： Qin dynasty xinggong （imperial mobile palace）， volume Ⅰ ［M］. Beijing： Cultural Relics Press， 2010： 261-266.
[14]	华玉冰，杨荣昌.姜女石：秦行宫遗址发掘报告（下册）［M］.北京：文物出版社， 2010： 10-265.
	HUA Y B， YANG R C. Excavation report of the Jiangnüshi site： Qin dynasty xinggong （imperial mobile palace）， volume Ⅱ ［M］. Beijing： Cultural Relics Press， 2010： 10-265.
[15]	中国社会科学院考古研究所，日本奈良国立文化财研究所.汉长安城桂宫： 1996 — 2001年考古发掘报告［M］.北京：文物出版社， 2007： 20-191.
	Institute of Archaeology of Chinese Academy of Social Sciences， Nara National Research Institute for Cultural Properties. The Guigong palace in the Han capital city Chang’an： archaeological excavation report from 1996 to 2001 ［M］. Beijing： Cultural Relics Press， 2007： 20-191.
[16]	中国社会科学院考古研究所.汉长安城武库：中国田野考古报告集［M］.北京：文物出版社， 2005： 61-68.
	Institute of Archaeology of Chinese Academy of Social Sciences. Armory in Chang’an city of the Han period： collection of reports on field archaeology in China ［M］. Beijing： Cultural Relics Press， 2005： 61-68.
[17]	中国社会科学院考古研究所.西汉礼制建筑遗址：考古学专刊［M］.北京：文物出版社， 2003： 166-196.
	Institute of Archaeology of Chinese Academy of Social Sciences. Ritual architecture sites of the Western Han dynasty： archaeology special issue ［M］. Beijing： Cultural Relics Press， 2003： 166-196.
[18]	中国社会科学院考古研究所.北魏洛阳永宁寺： 1979—1994年考古发掘报告［M］.北京：中国大百科全书出版社， 1996： 119-121.
	Institute of Archaeology of Chinese Academy of Social Sciences. The Yongningsi temple in Northern Wei： Luoyang excavations in 1979 — 1994 ［M］. Beijing： Encyclopedia of China Publishing House， 1996： 119-121.
[19]	中国社会科学院考古研究所.宁夏灵武窑发掘报告［M］.北京：中国大百科全书出版社， 1995： 83-84.
	Institute of Archaeology of Chinese Academy of Social Sciences. Excavations of the Lingwu kiln-site in Ningxia ［M］. Beijing： Encyclopedia of China Publishing House， 1995： 83-84.
[20]	辽宁省文物考古研究所，朝阳市北塔博物馆.朝阳北塔：考古发掘与维修工程报告［M］.北京：文物出版社， 2007： 22-61.
	Liaoning Provincial Institute of Cultural Relics and Archaeology， Chaoyang City North Pagoda Museum. Archaeological excavation and restoration project report of Chaoyang North Pagoda ［M］. Beijing： Cultural Relics Press， 2007： 22-61.
[21]	中国社会科学院考古研究所.隋仁寿宫·唐九成宫：考古发掘报告［M］.北京：科学出版社， 2008： 24-70.
	Institute of Archaeology of Chinese Academy of Social Sciences. Archaeological excavation report of the Renshou Palace （Sui dynasty） and Jiucheng Palace （Tang dynasty）［M］. Beijing： Science Press， 2008： 24-70.
[22]	洛阳市文物考古研究院.隋唐洛阳城天堂遗址发掘报告［M］.北京：科学出版社， 2017： 90-99.
	Luoyang Municipal Institute of Cultural Relics and Archaeology. Archaeological excavation report of the Tiantang site in the Sui-Tang imperial city of Luoyang ［M］. Beijing： Science Press， 2017： 90-99.
[23]	DAI Q， GAO Y. Hypergraph computation， AIFTA ［M］. Singapore： Springer， 2023： 1-8.
[24]	LIU Z， MAO H， WU C Y， et al. A ConvNet for the 2020s ［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 11966-11976.
[25]	GAO Y， ZHANG Z， LIN H， et al. Hypergraph learning： methods and practices ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2022， 44（5）： 2548-2566.
[26]	VOMMI A M， TIRUMALA KRISHNA B T. A hybrid filter-wrapper feature selection using fuzzy KNN based on Bonferroni mean for medical datasets classification： a COVID-19 case study ［J］. Expert Systems with Applications， 2023， 218： No.119612.
[27]	HUANG J， YANG J. UniGNN： a unified framework for graph and hypergraph neural networks ［C］// Proceedings of the 30th International Joint Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2021： 2563-2569.
[28]	XU K， HU W， LESKOVEC J， et al. How powerful are graph neural networks？［EB/OL］. ［2024-11-13］. .
[29]	DUGHMI S， KALAYCI Y， YORK G. Is transductive learning equivalent to PAC learning？［C］// Proceedings of the 36th International Conference on Algorithmic Learning Theory. New York： JMLR.org， 2025： 418-443.
[30]	DOSOVITSKIY A， BEYER L， KOLESNIKOV A， et al. An image is worth 16×16 words： Transformers for image recognition at scale ［EB/OL］. ［2024-11-13］. .
[31]	DAI Z， LIU H， LE Q V， et al. CoAtNet： marrying convolution and attention for all data sizes ［C］// Proceedings of the 35th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2021： 3965-3977.
[32]	LIU X， PENG H， ZHENG N， et al. EfficientViT： memory efficient vision Transformer with cascaded group attention ［C］// Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2023： 14420-14430.
[33]	FENG Y， YOU H， ZHANG Z， et al. Hypergraph neural networks ［C］// Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2019： 3558-3565.
[34]	DONG Y， SAWIN W， BENGIO Y. HNHN： hypergraph networks with hyperedge neurons ［EB/OL］. ［2024-11-13］. .
[35]	GAO Y， FENG Y， JI S， et al. HGNN⁺： general hypergraph neural networks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2023， 45（3）： 3181-3199.
[36]	JU W， MAO Z， YI S， et al. Hypergraph-enhanced dual semi-supervised graph classification ［C］// Proceedings of the 41st International Conference on Machine Learning. New York： JMLR.org， 2024： 22594-22604.
[37]	WANG S， ZHANG D W， HUANG J H， et al. Ada-HGNN： adaptive sampling for scalable hypergraph neural networks ［EB/OL］. ［2025-04-08］. .
[38]	MA X， ZHAO C， SHAO M， et al. Hypergraph-based dynamic graph node classification ［C］// Proceedings of the 2025 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2025： 1-5.

时期	瓦当样本数	时期	瓦当样本数
秦	128	北魏时期	28
汉	36	隋唐时期	117
西汉时期	136	辽	4
西汉中晚期	159	西夏时期	2
三燕时期	2

时期	瓦当样本数	时期	瓦当样本数
秦	128	北魏时期	28
汉	36	隋唐时期	117
西汉时期	136	辽	4
西汉中晚期	159	西夏时期	2
三燕时期	2

模型	准确率	加权精确率	加权召回率	加权F1-score
ResNet^［2］	73.66	70.69	73.66	68.19
VIT^［30］	89.98	89.36	89.98	88.97
ConvNeXt^［24］	87.41	86.86	87.41	86.85
CoAtNet^［31］	90.68	90.06	90.68	90.04
EfficientViT^［32］	87.88	87.82	87.88	85.66
UniGIN^［27］	84.05	84.83	84.05	83.83
HETD-DIC	91.36	92.12	91.36	91.38

模型	准确率	加权精确率	加权召回率	加权F1-score
ResNet^［2］	73.66	70.69	73.66	68.19
VIT^［30］	89.98	89.36	89.98	88.97
ConvNeXt^［24］	87.41	86.86	87.41	86.85
CoAtNet^［31］	90.68	90.06	90.68	90.04
EfficientViT^［32］	87.88	87.82	87.88	85.66
UniGIN^［27］	84.05	84.83	84.05	83.83
HETD-DIC	91.36	92.12	91.36	91.38

模型	准确率	加权精确率	加权召回率	加权F1-score
ResNet^［2］	84.55	79.61	84.55	80.45
VIT^［30］	90.24	89.74	90.24	89.85
ConvNeXt^［24］	87.80	88.69	87.80	87.79
CoAtNet^［31］	92.68	91.49	92.68	92.02
EfficientViT^［32］	90.24	89.38	90.24	87.99
UniGIN^［27］	88.35	86.23	88.35	86.62
HETD-DIC	91.65	89.54	91.65	90.09

Hypergraph-based eaves tile dating method under data imbalance conditions

基于超图的数据不平衡条件下的瓦当年代判别方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 14

References 38

Related Articles 15

Recommended Articles

Metrics

模型	准确率	加权精确率	加权召回率	加权F1-score
HGNN^［33］	87.01	87.08	87.01	85.87
HNHN^［34］	86.25	84.98	86.25	84.97
HGNN^+［35］	86.86	86.81	86.86	85.62
HEAL^［36］	74.64	77.26	74.64	75.29
Ada-HGNN^［37］	85.74	85.30	85.74	84.82
HYDG^［38］	76.99	79.94	76.99	77.94
UniGIN^［27］	90.08	90.35	90.08	89.50
HETD-DIC	94.75	94.90	94.75	94.59

模型	准确率	加权精确率	加权召回率	加权F1-score
HGNN^［33］	83.64	81.84	83.64	82.53
HNHN^［34］	83.06	81.29	83.06	81.87
HGNN^+［35］	82.40	80.35	82.40	81.29
HEAL^［36］	80.58	82.94	80.58	80.02
Ada-HGNN^［37］	90.74	91.06	90.74	90.47
HYDG^［38］	86.78	86.43	86.78	85.99
UniGIN^［27］	86.45	85.69	86.45	85.82
HETD-DIC	96.78	95.68	96.78	96.12

模型	准确率	加权精确率	加权召回率	加权F1-score
模型Ⅰ	90.08	90.35	90.08	89.50
模型Ⅱ	94.64	94.77	94.64	94.39
模型Ⅲ	94.28	94.30	94.28	94.03
模型Ⅳ	89.88	89.83	89.88	89.15
模型Ⅴ	89.47	89.36	89.47	88.75
HETD-DIC	94.75	94.90	94.75	94.59

[1]	Yanqun LU, Yiyi ZHAO. Customer churn prediction model integrating hierarchical graph neural network and specific feature learning [J]. Journal of Computer Applications, 2025, 45(9): 3057-3066.
[2]	Cong WANG, Yancui SHI. Group recommendation model by graph neural network based on multi-perspective learning [J]. Journal of Computer Applications, 2025, 45(4): 1205-1212.
[3]	Haijun GENG, Yun DONG, Zhiguo HU, Haotian CHI, Jing YANG, Xia YIN. Encrypted traffic classification method based on Attention-1DCNN-CE [J]. Journal of Computer Applications, 2025, 45(3): 872-882.
[4]	Hongyan ZHAO, Lihua GUO, Chunxia LIU, Riyun WANG. Scientific document summarization model based on multi-graph neural network and graph contrastive learning [J]. Journal of Computer Applications, 2025, 45(12): 3820-3828.
[5]	Jinghua ZHAO, Zhu ZHANG, Xiting LYU, Huidan LIN. Multiscale information diffusion prediction model based on hypergraph neural network [J]. Journal of Computer Applications, 2025, 45(11): 3529-3539.
[6]	Wenbo ZHAO, Zitong MA, Zhe YANG. Link prediction model based on directed hypergraph adaptive convolution [J]. Journal of Computer Applications, 2025, 45(1): 15-23.
[7]	Wenping ZHENG, Huilin GE, Meilin LIU, Gui YANG. Node classification algorithm fusing 2-connected motif-structure information [J]. Journal of Computer Applications, 2024, 44(5): 1464-1470.
[8]	Bin XIAO, Yun GAN, Min WANG, Xingpeng ZHANG, Zhaoxing WANG. Network abnormal traffic detection based on port attention and convolutional block attention module [J]. Journal of Computer Applications, 2024, 44(4): 1027-1034.
[9]	Li ZENG, Jingru YANG, Gang HUANG, Xiang JING, Chaoran LUO. Survey on hypergraph application methods： issues， advances， and challenges [J]. Journal of Computer Applications, 2024, 44(11): 3315-3326.
[10]	Nengqiang XIANG, Xiaofei ZHU, Zhaoze GAO. Information diffusion prediction model of prototype-aware dual-channel graph convolutional neural network [J]. Journal of Computer Applications, 2024, 44(10): 3260-3266.
[11]	Lantian XU, Ronghua LI, Yongheng DAI, Guoren WANG. Maximal clique searching algorithm for hypergraphs [J]. Journal of Computer Applications, 2023, 43(8): 2319-2324.
[12]	Weichao DANG, Bingyang CHENG, Gaimei GAO, Chunxia LIU. Contrastive hypergraph transformer for session-based recommendation [J]. Journal of Computer Applications, 2023, 43(12): 3683-3688.
[13]	Jing LIU, Zhihong DONG, Zheyu ZHANG, Zhigang SUN, Haipeng JI. Data sharing method of industrial internet of things based on federal incremental learning [J]. Journal of Computer Applications, 2022, 42(4): 1235-1243.
[14]	Xiaojie LI, Chaoran CUI, Guangle SONG, Yaxi SU, Tianze WU, Chunyun ZHANG. Stock trend prediction method based on temporal hypergraph convolutional neural network [J]. Journal of Computer Applications, 2022, 42(3): 797-803.
[15]	TIAN Ling, ZHANG Jinchuan, ZHANG Jinhao, ZHOU Wangtao, ZHOU Xue. Knowledge graph survey： representation， construction， reasoning and knowledge hypergraph theory [J]. Journal of Computer Applications, 2021, 41(8): 2161-2186.