Applied Intelligence最新文献

{"title":"DiffKT: a diffusion model for fine-grained knowledge tracing","authors":"Ruidi Liu,&nbsp;Yong Niu,&nbsp;Hu Li","doi":"10.1007/s10489-026-07459-9","DOIUrl":"10.1007/s10489-026-07459-9","url":null,"abstract":"<div><p>Accurately modeling learners’ evolving knowledge states is a long-standing challenge in intelligent education systems, especially under fine-grained temporal dynamics and inherently noisy learning interactions. Most existing knowledge tracing methods rely on deterministic graph-based or sequential modeling, which often struggle to disentangle genuine knowledge mastery from interaction noise such as guessing and careless errors. To address these challenges, this paper proposes DiffKT, a diffusion-based framework for fine-grained knowledge tracing that models learners’ knowledge states as probabilistic distributions rather than fixed point estimates. DiffKT integrates a dual-graph representation to capture student–question interactions and question–skill associations, together with a state-space sequence model that efficiently encodes long-range learning dependencies with linear complexity. Building upon these representations, a conditional diffusion model with an adaptive noise scheduling strategy is introduced to explicitly distinguish different types of interaction noise, enabling robust denoising and more accurate estimation of latent knowledge states. Extensive experiments on three real-world educational datasets demonstrate that DiffKT consistently outperforms advanced knowledge tracking methods in terms of prediction accuracy and stability, highlighting its effectiveness in modeling noisy and fine-grained learning behaviors.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ITCD: implicit trajectory-constrained diffusion for consistent time series imputation","authors":"Siyuan Ma,&nbsp;Ziwei Xu,&nbsp;Ryutaro Ichise","doi":"10.1007/s10489-026-07427-3","DOIUrl":"10.1007/s10489-026-07427-3","url":null,"abstract":"<div><p>Diffusion has recently attracted attention in time series imputation for its potential to model the uncertainty that deterministic methods often fail to capture. However, diffusion-based approaches often exhibit suboptimal imputation performance, as they overlook the distribution shift introduced by padding and rely on noise-driven optimization that ignores direct supervision signals from observed values. This causes the models to train on biased samples and struggle to approximate the observations, ultimately leading to inconsistent imputations. To address these limitations, we propose the <b>I</b>mplicit <b>T</b>rajectory-<b>C</b>onstrained <b>D</b>iffusion network (ITCD), which employs a two-stage diffusion architecture to mitigate the distribution shift caused by padding, thereby enabling better adaptation to realistic data. It also implicitly guides the diffusion sampling along a coherent denoising trajectory through intermediate and terminal constraints that consider both distribution and numerical accuracy, thereby achieving more consistent imputations. Extensive experiments illustrate that ITCD outperforms baselines across various scenarios with an average improvement of 13%, showing that it achieves more consistent time series imputation.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-026-07427-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal fault feature learning for machinery fault diagnosis by using recurrent attentional reinforcement learning","authors":"Zhenhui Tang,&nbsp;Jingcheng Wang,&nbsp;Shunyu Wu","doi":"10.1007/s10489-026-07434-4","DOIUrl":"10.1007/s10489-026-07434-4","url":null,"abstract":"<div><p>Data-driven fault diagnosis methods have demonstrated remarkable success by leveraging neural networks to automatically learn discriminative features from raw data. However, existing approaches often exhibit limited feature extraction capabilities, leading to suboptimal performance. To address this limitation, this paper proposes a novel fault diagnosis framework based on recurrent attentional reinforcement learning, which is designed to adaptively search optimal temporal features and identify fault types. The framework operates in three stages: firstly, a one-dimensional convolutional neural network (1D-CNN) extracts preliminary local temporal features. Secondly, a recurrent attentional module (RAM) iteratively localizes and samples the most informative temporal fragments. Finally, the attended fragments are fused to predict the health state of the machinery. A key advantage of the framework is its ability to search for informative fragments and explicitly model long-range dependencies across the selected temporal fragments, effectively capturing historical fault patterns. A reinforcement learning paradigm is introduced to optimize the entire feature extraction process in an end-to-end manner. The proposed method is evaluated on two mechanical systems, a rolling-bearing system and a hydraulic system, achieving average accuracies of 76.4% and 100.0%, respectively.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Traffic flow prediction using multiscale diffusion model guided by wavelet transform","authors":"Wenjing Wang,&nbsp;Deqian Fu","doi":"10.1007/s10489-026-07408-6","DOIUrl":"10.1007/s10489-026-07408-6","url":null,"abstract":"<div><p>Precise traffic flow forecasting is crucial for the development of intelligent transportation systems, enabling proactive traffic management, congestion alleviation, and efficient resource allocation. Traditional predictive models frequently fail to accurately capture the intricate nonlinear dynamics and multiscale temporal correlations present in traffic data. Multiscale Temporal- Spatial Diffusion Model informed by Wavelet Transform (MTSDM/WT) is proposed to overcome these constraints. The MTSDM/WT architecture begins with wavelet decomposition to decompose traffic time series into high-frequency components (capturing short-term fluctuations such as sudden congestion) and low-frequency components (reflecting long-term trends like rush-hour patterns). Specifically, the High-Frequency Refinement Module (HFRM) leverages multiscale convolution and attention mechanisms to precisely model localized temporal variations in the high-frequency domain, enhancing the capture of transient dynamics. Meanwhile, the low-frequency components undergo a forward diffusion process that incrementally adds noise, followed by a reverse denoising diffusion process to learn and reconstruct their temporal distribution, enabling accurate modeling of long-term trends. Finally, the refined high-frequency components and reconstructed low-frequency components are fused via inverse wavelet transform to generate comprehensive and robust traffic flow predictions. Extensive experiments were conducted on real-world traffic flow datasets from key urban intersections in Linyi City, Shandong Province, China. The MTSDM/WT framework exhibits notable improvements across critical evaluation metrics-Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Squared Error (RMSE), and Continuous Ranked Probability Score (CRPS)-with a 3% performance gain observed, for instance, on the MJ dataset at a 10-minute prediction horizon. These findings validate that MTSDM/WT provides superior prediction accuracy and resilience, especially in settings marked by fluctuating traffic patterns and noise disruption, with theoretical implications for advancing diffusion model applications in spatiotemporal series analysis and practical value for intelligent transportation system optimization.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ADPOD: An outlier detection algorithm based on affinity density peak clustering","authors":"Zhongping Zhang,&nbsp;Haoyu Tian,&nbsp;Xinlei Zhao,&nbsp;Guangxu Sun","doi":"10.1007/s10489-026-07430-8","DOIUrl":"10.1007/s10489-026-07430-8","url":null,"abstract":"<div><p>In recent years, outlier-detection approaches that integrate density estimation with clustering have received increasing attention, among which methods based on density-peak clustering (DPC) are the most prominent. Nevertheless, conventional DPC neglects local structural information when estimating density and still requires manual parameter tuning. Moreover, existing DPC-based outlier detectors lack an effective relative-distance mechanism for reliably discriminating normal instances from outliers. To address these limitations, this paper proposes an outlier detection algorithm based on affinity density peak clustering (ADPOD). First, an affinity density measure is defined based on both the k-nearest neighbor intersection size between samples and their corresponding distances. Second, cluster centers are automatically determined using the product of affinity density and decision distance. Third, a cluster relative distance is employed as the distance metric for data objects. Finally, a novel outlier factor is devised to quantify the outlier degree. Extensive experiments on both synthetic and real-world datasets demonstrate that the proposed algorithm significantly outperforms state-of-the-art competitors.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Remaining useful life prediction of electric vehicle drive system using physics-informed machine learning methods with uncertainty quantification","authors":"Zhen Wang,&nbsp;Zhengquan Chen,&nbsp;Wenqing Sun,&nbsp;Yadong Li,&nbsp;Yandong Hou,&nbsp;Lihui Zhao","doi":"10.1007/s10489-026-07429-1","DOIUrl":"10.1007/s10489-026-07429-1","url":null,"abstract":"<div><p>Electric vehicle drive system (EVDS) operates under complex multi-physical field coupling load conditions, posing significant challenges to modeling its performance degradation and assessing its remaining useful life (RUL). Accurate RUL prediction is essential for implementing predictive maintenance and enhancing driving safety. This study proposes an innovative physics-informed machine learning framework for RUL prediction and uncertainty quantification in EVDS. Initially, multidimensional time-domain and frequency-domain features are constructed based on actual operational load data, while multiscale cumulative damage features are developed by integrating the failure mechanisms of core components, followed by dimensionality reduction using the sparse autoencoder models. Subsequently, a RUL prediction model is developed by integrating Bayesian optimization with the convolutional neural network and bidirectional long short-term memory architecture, enhanced by an attention mechanism. To capture realistic degradation patterns of EVDS, accelerated durability bench tests are conducted, with the root mean square of vibration signals used to characterize degradation trajectories. By accounting for differences in degradation rates between user data and the accelerated test spectrum, nonlinear degradation trajectories for various users are generated. Finally, model training and parameter optimization are performed to predict RUL. The results demonstrate that RUL prediction errors remain within 5% across different service cycles. Compared to traditional degradation modeling-based prediction methods, the proposed approach effectively integrates physical information with deep learning algorithms, yielding a more concentrated probability density distribution of RUL predictions with higher accuracy, and enhanced generalization performance.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 14","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-grained power transformer fault diagnosis based on enhanced features of dissolved gas","authors":"Chenxu Meng,&nbsp;Dehe Fan,&nbsp;Mianting Wu,&nbsp;Tianzuo Yu,&nbsp;Bowen Xue,&nbsp;Yunan Lu","doi":"10.1007/s10489-026-07424-6","DOIUrl":"10.1007/s10489-026-07424-6","url":null,"abstract":"<div><p>Power transformers are critical components in electrical grids, and their operational reliability directly affects grid stability and power quality. Dissolved Gas Analysis (DGA) is a widely adopted technique for transformer fault diagnosis by analyzing decomposition gases dissolved in insulating oil. Existing data-driven DGA methods typically treat fault categories independently, neglecting the inherent hierarchical relationships among different fault types and severity levels, which limits model generalization capability. To address this issue, we propose MeFD, a multi-grained power transformer fault diagnosis framework based on enhanced dissolved gas features. Multi-grained learning refers to learning representations at multiple semantic granularity levels simultaneously, enabling the model to exploit hierarchical relationships among labels. Specifically, MeFD organizes transformer faults into a two-level hierarchy consisting of coarse-grained fault categories (normal, overheating, and discharging) and fine-grained severity levels. The diagnosis task is accordingly decomposed into coarse-grained fault classification and fine-grained severity regression. Furthermore, two discriminative feature enhancement strategies are introduced: (1) relative concentration ratios to characterize inter-gas relationships for fault classification, and (2) gas concentration deviations to quantify abnormality magnitude for severity regression. Experimental results on real-world datasets demonstrate that MeFD consistently improves the performance of multiple baseline machine learning algorithms and achieves superior accuracy in both fault recognition and severity regression.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-026-07424-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148838376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MD-VAE: Concepts embedded variational autoencoder with multiple decoders","authors":"Zitu Liu,&nbsp;Yue Liu,&nbsp;Shuang Li,&nbsp;Xing Wu,&nbsp;Yike Guo,&nbsp;Qun Liu,&nbsp;Guoyin Wang","doi":"10.1007/s10489-026-07435-3","DOIUrl":"10.1007/s10489-026-07435-3","url":null,"abstract":"<div><p>Variational autoencoders (VAE) construct latent space by optimizing the prior distribution and posterior distribution of the model. Existing methods exhibit limited interpretability during the construction of the latent space, which hinders their capacity to effectively capture the disentangled representation of concepts. To construct an interpretable latent space, we propose the Multi-Decoder Concept Embedding Variational Autoencoder (MD-VAE), which enhances latent space interpretability by learning distinct latent variables through multiple decoders. Firstly, the MD-VAE model learns prior concept by training on generated data that represent this concept, thereby embedding the prior into the latent space. Subsequently, we propose a variational inference framework utilizing multiple decoders. In this framework, encoders map multiple latent variables into the latent space, and each corresponding set of latent variables is reconstructed by its dedicated decoder. On the basis of this, a theoretical derivation of variational lower bound of multiple decodes is combined with variation method to obtain the optimal model parameter estimates. Finally, experiments on MNIST, FashionMNIST, COIL20, and USPS datasets show that MD-VAE can improve the prediction performance of VAE while discovering differences between different concepts.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148838249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hi-SpectralFormer: hierarchical spectral attention transformer for accurate medical image segmentation","authors":"Hao Xu,&nbsp;XiangJun Shen,&nbsp;Zhifeng Liu,&nbsp;Ming Yang,&nbsp;WenTao Kong,&nbsp;XiaoQin Qian","doi":"10.1007/s10489-026-07396-7","DOIUrl":"10.1007/s10489-026-07396-7","url":null,"abstract":"<div><p>Segmentation of medical images is still a difficult task, especially for high resolution images with complex anatomical variations that demand precise structure and boundary localization and effective global context modeling for accurate delineation of anatomical and pathological structure. While self-attention is effective in capturing long-range dependencies, it suffers from quadratic computational complexity, and local patch-based attention can break spatial continuity across regions and hinder making boundary-consistent structural modeling. In this work, we present a Hierarchical Spectral Attention Module which combines spectral linear attention with a frequency-aware pyramid to tackle these challenges. The proposed module keeps the geometric information related to the boundary while maintaining linear computational complexity by embedding graph-Laplacian spectral prior into the linear attention. The frequency-aware pyramid additionally makes a decomposition of features into various frequency components, which allows for the simultaneous modeling of global semantics, local textures, and fine-grained boundaries. Experimental results on ISIC 2018, Synapse 2D, and BraTS 2017 3D show that the proposed method achieves an improvement of 1.96% average Dice and Hausdorff distance reduction of 26.65% on Synapse, outperforming existing methods in terms of quality of both segmentation and boundary preservation.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148838363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Combining transformer global and task-relevant local CNN features for retail product recognition","authors":"Abdulaziz Ali,&nbsp;Siyuan You,&nbsp;Li Li,&nbsp;Gaoming He,&nbsp;Jian Yao","doi":"10.1007/s10489-026-07417-5","DOIUrl":"10.1007/s10489-026-07417-5","url":null,"abstract":"<div><p>Product recognition is a crucial task in computer vision and a fundamental component of retail systems, yet it remains challenging. Existing CNN-based approaches struggle to model global correlations effectively, while recent Transformer-based methods, despite their ability to capture long-range dependencies, overlook local interactions and fail to focus on task-relevant regions. To address these limitations, we propose a hybrid CNN-Transformer architecture that leverages the complementary strengths of both paradigms. Inspired by human visual perception, where individuals first perceive an object globally and then focus on salient local details, our method captures global representations and employs attention maps to refine task-relevant local features. First, Attentional Feature Fusion (AFF) and Stages Feature Fusion (SFF) modules are employed to extract robust CNN features by aggregating and integrating information across multiple scales. Second, the fused CNN features are fed into the Transformer, which leverages its self-attention mechanism; the multi-layer classification tokens [CLS] are integrated to generate the final global representation. Additionally, fused attention maps from multiple Transformer layers are used to filter CNN local features from only task-relevant regions, enabling the model to capture significant patterns while suppressing irrelevant noise. Finally, the concatenated global representations and refined local features are passed through ArcFace for classification. Extensive experiments on three benchmarks, Products-10K, RP2K, and Retail1K, demonstrate the effectiveness and superiority of the proposed framework over recent state-of-the-art methods.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148837742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}