Long time series prediction based on hybrid self-attention and differentiated normalization

doi:10.11772/j.issn.1001-9081.2025050628

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (5): 1499-1506.DOI: 10.11772/j.issn.1001-9081.2025050628

• Data science and technology • Previous Articles

Long time series prediction based on hybrid self-attention and differentiated normalization

Ruirui SONG, Leichun WANG(), Yunping HE, Jinxiang WEI, Xiangfeng LU, Xiaomeng LIU

School of Computer Science，Huber University，Wuhan Hubei 430062，China

Received:2025-06-06 Revised:2025-10-21 Accepted:2025-11-04 Online:2025-11-12 Published:2026-05-10
Contact: Leichun WANG
About author:SONG Ruirui， born in 1999， M. S. candidate. Her research interests include time series prediction， deep learning.
HE Yunping， born in 2000， M. S. candidate. His research interests include multimodal sentiment analysis， time series.
WEI Jinxiang， born in 2000， M. S. candidate. Her research interests include deep learning， spatio-temporal data prediction.
LU Xiangfeng， born in 2000， M. S. candidate. Her research interests include multimodal classification detection， deep learning.
LIU Xiaomeng， born in 2001， M. S. candidate. Her research interests include deep learning， multimodality.
Supported by:
National Natural Science Foundation of China(62106069);National Social Science Foundation of China(24BTQ019)

基于混合自注意力和差异归一化的长时间序列预测

宋芮芮, 王雷春(), 何运平, 魏金香, 卢祥凤, 刘小萌

湖北大学计算机学院，武汉 430062

通讯作者: 王雷春
作者简介:宋芮芮（1999—），女，山东枣庄人，硕士研究生，主要研究方向：时间序列预测、深度学习
何运平（2000—），男，湖北荆州人，硕士研究生，主要研究方向：多模态情感分析、时间序列
魏金香（2000—），女，安徽阜阳人，硕士研究生，主要研究方向：深度学习、时空数据预测
卢祥凤（2000—），女，山东临沂人，硕士研究生，主要研究方向：多模态分类检测、深度学习
刘小萌（2001—），女，山东枣庄人，硕士研究生，主要研究方向：深度学习、多模态。
基金资助:
国家自然科学基金资助项目(62106069);国家自然科学基金资助项目(62102136);国家社会科学基金资助项目(24BTQ019)

Abstract

Abstract:

Aiming at the problems of error accumulation， modeling difficulty， and low computational efficiency in long time series prediction， a long time series prediction model based on hybrid self-attention and differentiated normalization， namely HSADN （Hybrid Self-Attention and Differentiated Normalization）， was proposed. Firstly， the model used a stacked multi-head self-attention mechanism in the encoder to capture long-distance dependencies in time series， thereby reducing computational complexity， and used a multi-layer sparse self-attention mechanism in the decoder to dynamically adjust the generation strategy. Secondly， in the encoder， Batch Channel Normalization （BCN） was used to extract， fuse， and reconstruct the features， while in the decoder， Layer Normalization （LN） was adopted to alleviate the gradient vanishing and improve the training stability， generating predicted sequence values. Experimental results show that compared with CALF （Cross-modAl Large Language Model Fine-tuning） model， HSADN has the Mean Squared Error （MSE） and Mean Absolute Error （MAE） of univariate prediction reduced by 6.2% and 6.9% on ECL-960， respectively， and by 13.1% and 2.9% on ETTh-720， respectively； the MSE and MAE of multivariate prediction reduced by 3.5% and 2.6% on ETTm-672， respectively， and by 1.8% and 0.9% on Weather-720， respectively； the running time for univariate and multivariate predictions reduced by an average of 4.6% and 28.7%， respectively.

Key words: long time series prediction, self-attention mechanism, normalization, encoder, decoder, robustness

摘要：

针对长时间序列预测中存在的误差积累、建模困难和计算效率低的问题，提出一种基于混合自注意力和差异归一化的长时间序列预测模型HSADN（Hybrid Self-Attention and Differentiated Normalization）。首先，模型使用堆叠多头自注意力机制捕捉编码器中时间序列的长距离依赖，降低计算复杂度，并使用多层稀疏自注意力机制动态调整解码器中的生成策略；其次，在编码器中通过批量通道归一化（BCN）对特征进行提取、融合和重构，在解码器中通过层归一化（LN）缓解梯度消失和提升训练稳定性，输出预测序列值。实验结果表明，与CALF（Cross-modAl Large Language Model Fine-tuning）模型相比，HSADN的单变量预测的均方误差（MSE）与平均绝对误差（MAE）在ECL-960上分别降低6.2％和6.9%，在ETTh-720上分别降低13.1％和2.9%；多变量预测的MSE和MAE在ETTm-672上分别降低3.5％和2.6%，在Weather-720上分别降低1.8％和0.9%；在单变量和多变量预测时的运行时间分别平均降低4.6%和28.7%。

关键词: 长时间序列预测, 自注意力机制, 归一化, 编码器, 解码器, 鲁棒性

CLC Number:

TP391

Ruirui SONG, Leichun WANG, Yunping HE, Jinxiang WEI, Xiangfeng LU, Xiaomeng LIU. Long time series prediction based on hybrid self-attention and differentiated normalization[J]. Journal of Computer Applications, 2026, 46(5): 1499-1506.

宋芮芮, 王雷春, 何运平, 魏金香, 卢祥凤, 刘小萌. 基于混合自注意力和差异归一化的长时间序列预测[J]. 《计算机应用》唯一官方网站, 2026, 46(5): 1499-1506.

Figures/Tables 7

References 29

[1]	TORRES J F， MARTINEZ-BALLESTEROS M， TRONCOSO A， et al. Advances in time series forecasting： innovative methods and applications［J］. AIMS Mathematics， 2024， 9（9）： 24163-24165.
[2]	HYNDMAN R J， ATHANASOPOULOS G. Forecasting： principles and practice［M］. 3rd ed. Melbourne： OTexts， 2021： 145-177.
[3]	BENIDIS K， RANGAPURAM S S， FLUNKERT V， et al. Deep learning for time series forecasting： tutorial and literature survey［J］. ACM Computing Surveys， 2023， 55（6）： No.121.
[4]	LIM B， ZOHREN S. Time-series forecasting with deep learning： a survey［J］. Philosophical Transactions of the Royal Society A， 2021， 379（2194）： No.20200209.
[5]	KHALED A. BCN： batch channel normalization for image classification［C］// Proceedings of the 2024 International Conference on Pattern Recognition， LNCS 15311. Cham： Springer， 2025： 295-308.
[6]	MEHDIZADEH S. Using AR， MA， and ARMA time series models to improve the performance of MARS and KNN approaches in monthly precipitation modeling under limited climatic data［J］. Water Resources Management， 2020， 34（1）： 263-282.
[7]	ATESONGUN A， GULSEN M. A hybrid forecasting structure based on ARIMA and artificial neural network models［J］. Applied Sciences， 2024， 14（16）： No.7122.
[8]	ENGLE R. New frontiers for ARCH models［J］. Journal of Applied Econometrics， 2002， 17（5）： 425-446.
[9]	LARA-BENÍTEZ P， CARRANZA-GARCÍA M， RIQUELME J C. An experimental review on deep learning architectures for time series forecasting［J］. International Journal of Neural Systems， 2021， 31（3）： No.2130001.
[10]	SUN S， WANG L， LIN J， et al. An effective prediction model based on XGBoost for the 12-month recurrence of AF patients after RFA［J］. BMC Cardiovascular Disorders， 2023， 23： No.561.
[11]	KE G， 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. Red Hook： Curran Associates Inc.， 2017： 3149-3157.
[12]	SUCIATI I， USMAN M. Bayesian structural time series model for forecasting the composite stock price index in Indonesia［J］. Sciencestatistics： Journal of Statistics， Probability， and Its Application， 2023， 1（2）： 74-83.
[13]	TAYLOR S J， LETHAM B. Forecasting at scale［J］. The American Statistician， 2018， 72（1）： 37-45.
[14]	VAN DEN OORD A， DIELEMAN S， ZEN H， et al. WaveNet： a generative model for raw audio［EB/OL］. ［2025-01-23］. .
[15]	WAQAS M， HUMPHRIES U W. A critical review of RNN and LSTM variants in hydrological time series predictions［J］. MethodsX， 2024， 12（1）： No.102946.
[16]	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2017： 6000-6010.
[17]	ZHOU H， ZHANG S， PENG J， et al. Informer： beyond efficient Transformer for long sequence time-series forecasting［C］// Proceedings of the 35th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2021： 11106-11115.
[18]	WANG S， LI B Z， KHABSA M， et al. Linformer： self-attention with linear complexity［EB/OL］. ［2025-05-08］..
[19]	XIONG Y， ZENG Z， CHAKRABORTY R， et al. Nyströmformer： a Nyström-based algorithm for approximating self-attention［C］// Proceedings of the 35th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2021： 14138-14148.
[20]	LIN Y， KOPRINSKA I， RANA M. Temporal convolutional attention neural networks for time series forecasting［C］// Proceedings of the 2021 International Joint Conference on Neural Networks. Piscataway： IEEE， 2021： 1-8.
[21]	PATRO S G K， SAHU K K. Normalization： a preprocessing stage［J］. International Advanced Research Journal in Science， Engineering and Technology， 2015， 2（3）： 20-22.
[22]	HODSON T O. Root Mean Square Error （RMSE） or Mean Absolute Error （MAE）： when to use them or not［J］. Geoscientific Model Development， 2022， 15（14）： 5481-5487.
[23]	LI S， JIN X， XUAN Y， et al. Enhancing the locality and breaking the memory bottleneck of Transformer on time series forecasting［C］// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2019： 5243-5253.
[24]	KITAEV N， KAISER Ł， LEVSKAYA A. Reformer： the efficient Transformer［EB/OL］. ［2025-05-08］..
[25]	BAHDANAU D， CHO K， BENGIO Y. Neural machine translation by jointly learning to align and translate［EB/OL］. ［2025-05-08］..
[26]	SALINAS D， FLUNKERT V， GASTHAUS J， et al. DeepAR： probabilistic forecasting with autoregressive recurrent networks［J］. International Journal of Forecasting， 2020， 36（3）： 1181-1191.
[27]	LIU P， GUO H， DAI T， et al. CALF： aligning LLMs for time series forecasting via cross-modal fine-tuning［C］// Proceedings of the 39th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2025： 18915-18923.
[28]	LAI G， CHANG W， YANG Y， et al. Modeling long- and short-term temporal patterns with deep neural networks［C］// Proceedings of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval. New York： ACM， 2018： 95-104.
[29]	LIN S， LIN W， HU X， et al. CycleNet： enhancing time series forecasting through modeling periodic patterns［C］// Proceedings of the 38th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2024： 106315-106345.

数据集	预测步长	HSADN		CALF		CycleNet		Informer		LogTrans		Reformer		LSTMa		DeepAR
数据集	预测步长	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE
ETTh	168	0.158	0.325	0.161	0.329	0.173	0.335	0.183	0.346	0.207	0.375	1.522	1.191	0.236	0.392	0.239	0.422
	336	0.169	0.339	0.175	0.357	0.217	0.361	0.222	0.387	0.230	0.398	1.860	1.124	0.590	0.698	0.445	0.552
	720	0.192	0.363	0.221	0.374	0.245	0.385	0.269	0.435	0.273	0.463	2.112	1.436	0.683	0.768	0.658	0.707
ETTm	96	0.106	0.266	0.145	0.278	0.141	0.269	0.194	0.372	0.199	0.386	0.920	0.767	0.287	0.420	0.364	0.496
	288	0.278	0.440	0.324	0.459	0.364	0.467	0.401	0.554	0.411	0.572	1.108	1.245	0.524	0.584	0.948	0.795
	672	0.345	0.502	0.371	0.504	0.425	0.511	0.512	0.644	0.598	0.702	1.793	1.528	1.064	0.873	2.437	1.352
Weather	168	0.206	0.330	0.220	0.334	0.232	0.342	0.266	0.398	0.309	0.439	0.654	0.634	0.341	0.448	0.293	0.451
	336	0.214	0.338	0.236	0.343	0.278	0.351	0.297	0.416	0.359	0.484	1.792	1.093	0.456	0.554	0.585	0.644
	720	0.245	0.369	0.267	0.375	0.323	0.382	0.359	0.466	0.388	0.499	2.087	1.534	0.866	0.809	0.499	0.596
ECL	168	0.376	0.480	0.369	0.455	0.419	0.468	0.447	0.503	0.454	0.529	1.671	1.587	0.723	0.655	0.315	0.436
	336	0.388	0.486	0.396	0.487	0.474	0.497	0.501	0.536	0.514	0.563	3.528	2.196	1.212	0.898	0.414	0.519
	720	0.449	0.539	0.463	0.542	0.506	0.547	0.552	0.577	0.558	0.609	4.891	4.047	1.511	0.966	0.563	0.595
	960	0.481	0.502	0.513	0.539	0.544	0.566	0.582	0.608	0.624	0.645	7.019	5.105	1.545	1.006	0.657	0.683
Count		24		0		0		0		0		0		0		2

数据集	预测步长	HSADN		CALF		CycleNet		Informer		LogTrans		Reformer		LSTMa		DeepAR
数据集	预测步长	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE
ETTh	168	0.158	0.325	0.161	0.329	0.173	0.335	0.183	0.346	0.207	0.375	1.522	1.191	0.236	0.392	0.239	0.422
	336	0.169	0.339	0.175	0.357	0.217	0.361	0.222	0.387	0.230	0.398	1.860	1.124	0.590	0.698	0.445	0.552
	720	0.192	0.363	0.221	0.374	0.245	0.385	0.269	0.435	0.273	0.463	2.112	1.436	0.683	0.768	0.658	0.707
ETTm	96	0.106	0.266	0.145	0.278	0.141	0.269	0.194	0.372	0.199	0.386	0.920	0.767	0.287	0.420	0.364	0.496
	288	0.278	0.440	0.324	0.459	0.364	0.467	0.401	0.554	0.411	0.572	1.108	1.245	0.524	0.584	0.948	0.795
	672	0.345	0.502	0.371	0.504	0.425	0.511	0.512	0.644	0.598	0.702	1.793	1.528	1.064	0.873	2.437	1.352
Weather	168	0.206	0.330	0.220	0.334	0.232	0.342	0.266	0.398	0.309	0.439	0.654	0.634	0.341	0.448	0.293	0.451
	336	0.214	0.338	0.236	0.343	0.278	0.351	0.297	0.416	0.359	0.484	1.792	1.093	0.456	0.554	0.585	0.644
	720	0.245	0.369	0.267	0.375	0.323	0.382	0.359	0.466	0.388	0.499	2.087	1.534	0.866	0.809	0.499	0.596
ECL	168	0.376	0.480	0.369	0.455	0.419	0.468	0.447	0.503	0.454	0.529	1.671	1.587	0.723	0.655	0.315	0.436
	336	0.388	0.486	0.396	0.487	0.474	0.497	0.501	0.536	0.514	0.563	3.528	2.196	1.212	0.898	0.414	0.519
	720	0.449	0.539	0.463	0.542	0.506	0.547	0.552	0.577	0.558	0.609	4.891	4.047	1.511	0.966	0.563	0.595
	960	0.481	0.502	0.513	0.539	0.544	0.566	0.582	0.608	0.624	0.645	7.019	5.105	1.545	1.006	0.657	0.683
Count		24		0		0		0		0		0		0		2

数据集	预测步长	HSADN		CALF		CycleNet		Informer		LogTrans		Reformer		LSTMa		LSTNet
数据集	预测步长	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE
ETTh	168	0.816	0.695	0.854	0.713	0.875	0.726	0.931	0.752	1.002	0.846	1.824	1.138	1.212	0.867	1.997	1.214
	336	0.987	0.778	1.059	0.791	1.107	0.809	1.128	0.873	1.362	0.952	2.117	1.280	1.424	0.994	2.655	1.369
	720	1.133	0.840	1.166	0.863	1.195	0.874	1.215	0.896	1.397	1.291	2.415	1.520	1.960	1.322	2.143	1.380
ETTm	96	0.537	0.524	0.552	0.551	0.548	0.556	0.678	0.614	0.768	0.792	1.433	0.945	1.339	0.913	2.762	1.542
	288	0.702	0.624	0.733	0.636	0.795	0.669	1.056	0.786	1.462	1.320	1.820	1.094	1.740	1.124	1.257	2.076
	672	0.806	0.683	0.835	0.701	0.865	0.713	1.192	0.926	1.669	1.461	2.187	1.232	2.736	1.555	1.917	2.941
Weather	168	0.561	0.545	0.544	0.528	0.595	0.587	0.608	0.567	0.727	0.671	1.318	0.855	1.038	0.835	0.748	0.647
	336	0.585	0.563	0.592	0.574	0.616	0.603	0.702	0.620	0.754	0.670	1.930	1.167	1.657	1.059	0.782	0.683
	720	0.593	0.568	0.604	0.573	0.625	0.611	0.831	0.731	0.885	0.773	2.726	1.575	1.536	1.109	0.851	0.757
ECL	168	0.319	0.405	0.303	0.385	0.332	0.408	0.374	0.427	0.368	0.432	1.515	1.069	0.574	0.602	0.394	0.476
	336	0.341	0.406	0.346	0.412	0.360	0.420	0.381	0.431	0.373	0.439	1.601	1.104	0.886	0.795	0.419	0.477
	720	0.361	0.427	0.385	0.434	0.402	0.447	0.426	0.454	0.409	0.454	2.009	1.170	1.676	1.095	0.556	0.565
	960	0.425	0.470	0.436	0.489	0.441	0.498	0.460	0.548	0.477	0.589	2.141	1.387	1.591	1.128	0.605	0.599
Count		22		4		0		0		0		0		0		0

数据集	预测步长	HSADN		CALF		CycleNet		Informer		LogTrans		Reformer		LSTMa		LSTNet
数据集	预测步长	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE	MSE	MAE
ETTh	168	0.816	0.695	0.854	0.713	0.875	0.726	0.931	0.752	1.002	0.846	1.824	1.138	1.212	0.867	1.997	1.214
	336	0.987	0.778	1.059	0.791	1.107	0.809	1.128	0.873	1.362	0.952	2.117	1.280	1.424	0.994	2.655	1.369
	720	1.133	0.840	1.166	0.863	1.195	0.874	1.215	0.896	1.397	1.291	2.415	1.520	1.960	1.322	2.143	1.380
ETTm	96	0.537	0.524	0.552	0.551	0.548	0.556	0.678	0.614	0.768	0.792	1.433	0.945	1.339	0.913	2.762	1.542
	288	0.702	0.624	0.733	0.636	0.795	0.669	1.056	0.786	1.462	1.320	1.820	1.094	1.740	1.124	1.257	2.076
	672	0.806	0.683	0.835	0.701	0.865	0.713	1.192	0.926	1.669	1.461	2.187	1.232	2.736	1.555	1.917	2.941
Weather	168	0.561	0.545	0.544	0.528	0.595	0.587	0.608	0.567	0.727	0.671	1.318	0.855	1.038	0.835	0.748	0.647
	336	0.585	0.563	0.592	0.574	0.616	0.603	0.702	0.620	0.754	0.670	1.930	1.167	1.657	1.059	0.782	0.683
	720	0.593	0.568	0.604	0.573	0.625	0.611	0.831	0.731	0.885	0.773	2.726	1.575	1.536	1.109	0.851	0.757
ECL	168	0.319	0.405	0.303	0.385	0.332	0.408	0.374	0.427	0.368	0.432	1.515	1.069	0.574	0.602	0.394	0.476
	336	0.341	0.406	0.346	0.412	0.360	0.420	0.381	0.431	0.373	0.439	1.601	1.104	0.886	0.795	0.419	0.477
	720	0.361	0.427	0.385	0.434	0.402	0.447	0.426	0.454	0.409	0.454	2.009	1.170	1.676	1.095	0.556	0.565
	960	0.425	0.470	0.436	0.489	0.441	0.498	0.460	0.548	0.477	0.589	2.141	1.387	1.591	1.128	0.605	0.599
Count		22		4		0		0		0		0		0		0

模型	评价指标	预测步长
模型	评价指标	168	336	720
HSADN-NA	MSE	0.165	0.181	0.221
HSADN-NA	MAE	0.327	0.340	0.384
HSADN-NN	MSE	0.162	0.181	0.212
HSADN-NN	MAE	0.326	0.350	0.378
HSADN-NAN	MSE	0.183	0.222	0.269
HSADN-NAN	MAE	0.346	0.387	0.435
HSADN	MSE	0.158	0.169	0.192
HSADN	MAE	0.325	0.339	0.363

Long time series prediction based on hybrid self-attention and differentiated normalization

基于混合自注意力和差异归一化的长时间序列预测

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 29

Related Articles 15

Recommended Articles

Metrics

[1]	Qianfei WANG, Yang LI, Deyu LI, Suge WANG. Dual-channel feature fusion representation method for short-text clustering based on large language model [J]. Journal of Computer Applications, 2026, 46(5): 1441-1449.
[2]	Jie HU, Pengcheng LI, Jun SUN, Jiaao ZHANG. Key phrase extraction model based on multi-perspective information enhancement and hierarchical weighting [J]. Journal of Computer Applications, 2026, 46(4): 1050-1057.
[3]	Hu LUO, Mingshu ZHANG. Rumor detection method based on cross-modal attention mechanism and contrastive learning [J]. Journal of Computer Applications, 2026, 46(2): 361-367.
[4]	Xiuyan ZHANG, Wentao LIU, Xin WANG. Feature extraction method of flight pitch operation based on quick access recorder data [J]. Journal of Computer Applications, 2026, 46(1): 322-330.
[5]	Xiang WANG, Zhixiang CHEN, Guojun MAO. Multivariate time series prediction method combining local and global correlation [J]. Journal of Computer Applications, 2025, 45(9): 2806-2816.
[6]	Dengran REN, Shuying WANG. Nested named entity recognition model for wind power equipment based on differential boundary enhancement [J]. Journal of Computer Applications, 2025, 45(9): 2798-2805.
[7]	Li LI, Han SONG, Peihe LIU, Hanlin CHEN. Named entity recognition for sensitive information based on data augmentation and residual networks [J]. Journal of Computer Applications, 2025, 45(9): 2790-2797.
[8]	Yiming LIANG, Jing FAN, Wenze CHAI. Multi-scale feature fusion sentiment classification based on bidirectional cross attention [J]. Journal of Computer Applications, 2025, 45(9): 2773-2782.
[9]	Ao SHEN, Ruizhang HUANG, Jingjing XUE, Yanping CHEN, Yongbin QIN. Deep variational text clustering model based on distribution augmentation [J]. Journal of Computer Applications, 2025, 45(8): 2457-2463.
[10]	Chen LIANG, Yisen WANG, Qiang WEI, Jiang DU. Source code vulnerability detection method based on Transformer-GCN [J]. Journal of Computer Applications, 2025, 45(7): 2296-2303.
[11]	Zhenzhou WANG, Fangfang GUO, Jingfang SU, He SU, Jianchao WANG. Robustness optimization method of visual model for intelligent inspection [J]. Journal of Computer Applications, 2025, 45(7): 2361-2368.
[12]	Xiaoqiang ZHAO, Yongyong LIU, Yongyong HUI, Kai LIU. Batch process quality prediction model using improved time-domain convolutional network with multi-head self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2245-2252.
[13]	Huibin WANG, Zhan’ao HU, Jie HU, Yuanwei XU, Bo WEN. Time series forecasting model based on segmented attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2262-2268.
[14]	Jie JIANG, Gongning LUO, Suyu DONG, Fanding LI, Xiangyu LI, Qince LI, Yongfeng YUAN, Kuanquan WANG. Information bottleneck-guided intracranial hemorrhage segmentation method [J]. Journal of Computer Applications, 2025, 45(6): 1998-2006.
[15]	Daoquan LI, Zheng XU, Sihui CHEN, Jiayu LIU. Network traffic classification model integrating variational autoencoder and AdaBoost-CNN [J]. Journal of Computer Applications, 2025, 45(6): 1841-1848.