Tabular Ordinal Classification Based on Ordinal Entropy Optimization Model

doi:10.19678/j.issn.1000-3428.0252092

Computer Engineering ›› 2026, Vol. 52 ›› Issue (9): 81-91. doi: 10.19678/j.issn.1000-3428.0252092

• Computational Intelligence and Pattern Recognition • Previous Articles Next Articles

Tabular Ordinal Classification Based on Ordinal Entropy Optimization Model

LUO Zhengdong¹^,²^,³, ZHANG Guohao¹^,²^,³, HAN Yunfei¹^,²^,³, WANG Yi¹^,²^,³, ZHOU Xi¹^,²^,³^,*()

1. Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, Xinjiang, China
2. University of Chinese Academy of Sciences, Beijing 100049, China
3. Xinjiang Key Laboratory of Minority Speech and Language Information Processing, Chinese Academy of Sciences, Urumqi 830011, Xinjiang, China

Received:2025-01-24 Revised:2025-03-14 Online:2026-09-15 Published:2025-05-09
Contact: ZHOU Xi

基于有序熵优化模型的表格有序分类

罗正东¹^,²^,³, 张国昊¹^,²^,³, 韩云飞¹^,²^,³, 王轶¹^,²^,³, 周喜¹^,²^,³^,*()

1. 中国科学院新疆理化技术研究所, 新疆乌鲁木齐 830011
2. 中国科学院大学, 北京 100049
3. 中国科学院新疆民族语音语言信息处理重点实验室, 新疆乌鲁木齐 830011

通讯作者: 周喜
作者简介:
罗正东, 男, 博士研究生, 主研方向为数据挖掘、表格有序分类
张国昊, 博士研究生
韩云飞(CCF会员), 副研究员、博士
王轶(CCF高级会员), 研究员、博士
周喜(通信作者), 研究员、博士
基金资助:
新疆维吾尔自治区重大科技专项(2023A01006); 新疆维吾尔自治区重点研发计划(2023B01028); 新疆维吾尔自治区"天山英才"项目(2022TSYCLJ0035); 新疆维吾尔自治区"天山英才"项目(2023TSYCTD0011); 新疆维吾尔自治区"天山英才"项目(2023TSYCCX0046); 中国科学院青年创新促进会项目(2021434)

Abstract

Abstract:

The existing methods for tabular data prediction primarily focus on classical classification and regression tasks. However, a type of data in the tabular data domain contains labels with an ordinal relationship, and its prediction task is called tabular ordinal classification. Current methods for tabular ordinal classification rely primarily on retrieving similar features and augmenting sample feature representations by fusing features similar to the ordinal distance between classes. However, the existing methods neglect the full utilization of label ordinal knowledge. To address this, a method based on ordinal label entropy optimization is proposed, which effectively guides the model in learning ordinal information by mining the ordinal entropy embedded in the label order knowledge. First, an ordinal entropy calculation module is established that quantifies ordinal entropy based on the rank differences between the predicted and true labels. Through step-by-step analysis and derivation, the ordinal label entropy is designed as a novel rank loss function, which is introduced as a regularization term in the model. This encourages the model to learn the ordinal relationship between labels and reduces information loss caused by unordered predictions. The ordinal entropy optimized ranking loss function is combined with the original loss function of the model to improve its predictive ability. Finally, experiments on multiple ordinal tabular datasets reveal that this method outperforms various baseline models, demonstrating the effectiveness and advantages of the ordinal entropy optimization model in tabular ordinal classification tasks.

Key words: tabular data, ordinal classification, ordinal information, ordinal entropy, rank loss

摘要：

现有表格数据预测方法主要聚焦于传统分类和回归的研究, 然而在表格数据领域中存在一种标签具有有序关系的数据类型, 其预测任务被称为表格有序分类。目前的表格有序分类方法主要采用检索相似特征的方式, 通过相似特征与类间有序距离融合增强样本特征表示。但现有方法忽略了标签有序知识的充分利用, 因此提出一种基于标签有序熵优化的方法, 通过挖掘标签有序知识中蕴含的有序熵, 有效指导模型学习有序信息。具体而言, 首先建立有序熵计算模块, 利用预测标签与真实标签之间的等级顺序差异量化有序熵。通过逐步分析和推导, 将标签有序熵设计为一种新颖的排序损失函数作为正则项引入模型, 鼓励模型学习标签等级顺序关系, 以减少无序预测带来的信息损失。然后, 将有序熵优化排序损失函数与模型原有损失函数相结合, 共同提升模型的预测能力。在多个有序表格数据集上的实验结果显示, 该方法相较于多种基线模型取得了性能提升, 充分证明了有序熵优化模型在表格有序分类任务中的有效性与优势。

关键词: 表格数据, 有序分类, 有序信息, 有序熵, 排序损失

LUO Zhengdong, ZHANG Guohao, HAN Yunfei, WANG Yi, ZHOU Xi. Tabular Ordinal Classification Based on Ordinal Entropy Optimization Model[J]. Computer Engineering, 2026, 52(9): 81-91.

罗正东, 张国昊, 韩云飞, 王轶, 周喜. 基于有序熵优化模型的表格有序分类[J]. 计算机工程, 2026, 52(9): 81-91.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0252092

https://www.ecice06.com/EN/Y2026/V52/I9/81

Figures/Tables 9

References 32

1	SHWARTZ-ZIV R , ARMON A . Tabular data: deep learning is not all you need. Information Fusion, 2022, 81, 84- 90. doi: 10.1016/j.inffus.2021.11.011
2	WU K , WU E , DANDREA M , et al. Machine learning prediction of clinical trial operational efficiency. The AAPS Journal, 2022, 24(3): 57. doi: 10.1208/s12248-022-00703-3
3	YAN J H, ZHENG B, XU H X, et al. Making pre-trained language models great on tabular prediction[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2403.01841.
4	贺超, 汪超凡, 张思睿, 等. 电力物联网自适应任务卸载策略. 重庆邮电大学学报(自然科学版), 2025, 37(4): 507- 515.
	HE C , WANG C F , ZHANG S R , et al. Adaptive task offloading strategy for power Internet of Things. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2025, 37(4): 507- 515.
5	常萨, 冯勇. 多模态深度学习的区块链智能合约漏洞检测方法. 小型微型计算机系统, 2025, 46(4): 958- 965.
	CHANG S , FENG Y . Multi-modal deep learning blockchain smart contract vulnerability detection method. Journal of Chinese Computer Systems, 2025, 46(4): 958- 965.
6	CHEN J T, YAN J H, CHEN Q Y, et al. ExcelFormer: a neural network surpassing GBDTs on tabular data[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2301.02819.
7	LUO Z D, ATAWULLA A, YANG F Y, et al. TabCGOK: intra-class groups retrieval and inter-class ordinal knowledge augmented network for ordinal tabular data prediction[C]//Proceedings of ECAI 2024. Santiago de Compostela, Spain: Springer Nature Switzerland, 2024: 2242-2249.
8	冉烔宇, 汤梦姿, 解庆, 等. 基于信息融合的半监督有序分类框架. 计算机工程, 2026, 52(2): 287- 298. doi: 10.19678/j.issn.1000-3428.0069814 URL
	RAN T Y , TANG M Z , XIE Q , et al. Semi-supervised ordinal classification framework based on information fusion. Computer Engineering, 2026, 52(2): 287- 298. doi: 10.19678/j.issn.1000-3428.0069814 URL
9	ZHA K W, CAO P, SON J, et al. Rank-N-contrast: learning continuous representations for regression[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2210.01189.
10	FANCONI C, VAN BUCHEM M, HERNANDEZ-BOUSSARD T. Natural language processing methods to identify oncology patients at high risk for acute care with clinical notes[EB/OL]. [2024-10-11]. http://arxiv.org/pdf/2209.13860.
11	FRANKS B J, DINKELMANN B, FELLENZ S, et al. Ordinal regression for difficulty estimation of StepMania levels[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2301.09485.
12	HUANG X, KHETAN A, CVITKOVIC M, et al. TabTransformer: tabular data modeling using contextual embeddings[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2012.06678.
13	SONG W P, SHI C C, XIAO Z P, et al. Autoint: automatic feature interaction learning via self-attentive neural networks[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. New York, USA: ACM Press, 2019: 1161-1170.
14	GORISHNIY Y, RUBACHEV I, KARTASHEV N, et al. TabR: tabular deep learning meets nearest neighbors in 2023[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2307.14338.
15	NADER Y, SIXT L, LANDGRAF T. DNNR: differential nearest neighbors regression[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2205.08434.
16	HU W H, YUAN Y W, ZHANG Z C, et al. PyTorch Frame: a modular framework for multi-modal tabular learning[EB/OL]. [2024-10-11]. https://arxiv.org/html/2404.00776v2.
17	FRIEDMAN J H . Greedy function approximation: a gradient boosting machine. The Annals of Statistics, 2001, 29(5): 1189- 1232. doi: 10.1214/aos/1013203450
18	CHEN T Q, GUESTRIN C. XGBoost: a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2016: 785-794.
19	KE G L, MENG Q, FINLEY T, et al. LightGBM: a highly efficient gradient boosting decision tree[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 3149-3157.
20	PROKHORENKOVA L, GUSEV G, VOROBEV A, et al. CatBoost: unbiased boosting with categorical features[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2018: 6639-6649.
21	VASWANI A. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 6000-6010.
22	ARIK S Ö, PFISTER T. TabNet: attentive interpretable tabular learning[C]// Proceedings of the AAAI Conference on Artificial Intelligence. [S.l.]: AAAI Press, 2021: 6679-6687.
23	SOMEPALLI G, GOLDBLUM M, SCHWARZSCHILD A, et al. SAINT: improved neural networks for tabular data via row attention and contrastive pre-training[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2106.01342.
24	KOSSEN J, BAND N, LYLE C, et al. Self-attention between datapoints: going beyond individual input-output pairs in deep learning[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2106.02584.
25	SHI X T , CAO W Z , RASCHKA S . Deep neural networks for rank-consistent ordinal regression based on conditional probabilities. Pattern Analysis and Applications, 2023, 26(3): 941- 955. doi: 10.1007/s10044-023-01181-9
26	SÁNCHEZ-MONEDERO J , GUTIÉRREZ P A , TIŇO P , et al. Exploitation of pairwise class distances for ordinal classification. Neural Computation, 2013, 25(9): 2450- 2485. doi: 10.1162/NECO_a_00478
27	TWOMEY N, POYIADZI R, MANN C, et al. Ordinal regression as structured classification[EB/OL]. [2024-10-11]. https://arxiv.org/abs/1905.13658.
28	FUCHS T S, KESHET J. THOR: threshold-based ranking loss for ordinal regression[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2205.04864.
29	GUTIÉRREZ P A , PÉREZ-ORTIZ M , SÁNCHEZ-MONEDERO J , et al. Ordinal regression methods: survey and experimental study. IEEE Transactions on Knowledge and Data Engineering, 2016, 28(1): 127- 146. doi: 10.1109/TKDE.2015.2457911
30	GORISHNIY Y, RUBACHEV I, KHRULKOV V, et al. Revisiting deep learning models for tabular data[C]//Proceedings of the 35th Annual Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2021: 18932-18943.
31	AHAMED M A, CHENG Q. MambaTab: a plug-and-play model for learning tabular data[C]//Proceedings of the IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR). San Jose, USA: IEEE Press, 2024: 369-375.
32	CHEN K Y, CHIANG P H, CHOU H R, et al. Trompt: towards a better deep neural network for tabular data[EB/OL]. [2024-10-11]. https://arxiv.org/abs/2305.18446.

[1]	RAN Tongyu, TANG Mengzi, XIE Qing, LIU Yongjian. Semi-Supervised Ordinal Classification Framework Based on Information Fusion [J]. Computer Engineering, 2026, 52(2): 287-298.
[2]	GUO Li-hong; WU Hai-tao. TDS Protocol Analysis and Loophole Detection [J]. Computer Engineering, 2009, 35(18): 127-129.

Please choose a citation manager

Content to export