Applied Intelligence最新文献

{"title":"FutuTP: Future-based trajectory prediction for autonomous driving","authors":"Qingchao Xu,&nbsp;Yandong Liu,&nbsp;Shixi Wen,&nbsp;Xin Yang,&nbsp;Dongsheng Zhou","doi":"10.1007/s10489-025-06510-5","DOIUrl":"10.1007/s10489-025-06510-5","url":null,"abstract":"<div><p>Trajectory prediction is an essential aspect of autonomous driving technology. Based on the historical trajectories and environmental information, trajectory prediction methods predict the future trajectory of a vehicle. Goal-based methods have been successful because of their excellent interpretability. However, these methods ignore future lane information and interactions in future trajectories, which leads to prediction failures in some scenes. In this paper, we propose an encoder-decoder model called future-based trajectory prediction (FutuTP). The encoder fuses the interactions of future trajectories through a transformer module. The decoder predicts the future lane area and applies the results to generate a trajectory. The experimental results show that FutuTP achieves more accurate predictions than does the SOTA method on Argoverse 1. Especially in terms of the <span>(text {minFDE}_6)</span> metric, FutuTP outperforms the SOTA method by approximately 6%. The code can be accessed via the following link: https://github.com/Qingchao-Xu/FutuTP.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761780","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":"DBFF-GRU: dual-branch temporal feature fusion network with fast GRU for multivariate time series forecasting","authors":"Jinglei Li,&nbsp;Dongsheng Liu,&nbsp;Guofang Ma,&nbsp;Yaning Chen,&nbsp;Hongwei Jiang","doi":"10.1007/s10489-025-06447-9","DOIUrl":"10.1007/s10489-025-06447-9","url":null,"abstract":"<div><p>Multivariate time series (MTS) forecasting involves the use of multiple interrelated sequential data to predict future trends, necessitating the extraction of potential associative information from complex historical data. Currently, Transformers dominate the field of MTS prediction due to their core mechanism of self-attention, which effectively captures long-range dependencies. However, self-attention is inherently permutation-invariant, leading to the loss of sequential information. To address this issue, we propose the Dual-Branch Temporal Feature Fusion Network with Fast GRU (DBFF-GRU). In the feature fusion module, a dual-branch convolutional structure is employed to extract local and global features from the time series data separately, and a lightweight attention module is integrated into the global feature branch to capture dependencies among variables. Additionally, we introduce a fast iterative GRU structure to further capture long-term dependencies and enhance model efficiency. Extensive experiments on real-world data demonstrate the effectiveness of DBFF-GRU compared to state-of-the-art techniques.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769823","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":"Dual-branch contrastive learning for weakly supervised object localization","authors":"Zebin Guo,&nbsp;Dong Li,&nbsp;Zhengjun Du,&nbsp;Bingfeng Seng","doi":"10.1007/s10489-025-06514-1","DOIUrl":"10.1007/s10489-025-06514-1","url":null,"abstract":"<div><p>The weakly supervised object localization task uses image-level labels to train object localization models. Traditional convolutional neural network (CNN)-based methods usually localize objects using a class activation map. However, the class activation map usually suffers from the problem of activating a small part of the object that is most discriminative. Meanwhile, the methods based on the Vision Transformer can capture long-range feature dependencies but tend to ignore local feature details. In this paper, we innovatively propose a dual-branch contrastive learning (DBC) method that consists of a Transformer and a CNN branch. The method can effectively separate the background and foreground of an image and fuse the features of Transformer and CNN through contrastive learning. Specifically, the method separates the background and foreground representations of the image using the initially generated class-agnostic activation maps. Then, the representations of the same image from different branches form positive pairs for contrastive learning. The background and foreground representations from the same branch form negative pairs. Finally, the DBC method forces the model to separate the background and foreground representations through negative contrastive loss and makes the model fuse the features of two branches through positive contrastive loss. Experiments on the ILSVRC benchmark show that the proposed method can achieve a Top-1 localization accuracy of 59.9% and a GT-known localization accuracy of 71.7%, which are better metrics than those of the state-of-the-art methods with the same parameter complexity.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761740","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":"Global and local co-attention networks enhanced by learning state for knowledge tracing","authors":"Xinhua Wang,&nbsp;Yibang Cao,&nbsp;Liancheng Xu,&nbsp;Ke Sun","doi":"10.1007/s10489-025-06463-9","DOIUrl":"10.1007/s10489-025-06463-9","url":null,"abstract":"<div><p>In intelligent tutoring systems, knowledge tracing (KT) stands as a pivotal technology for facilitating personalized learning among students. Effectively capturing the continually evolving knowledge mastery states of students poses a formidable challenge in KT prediction. Traditional KT methods typically model students’ global knowledge mastery states solely based on the chronological sequence of their historical interactions, neglecting the significance of their current learning state and the inherent interplay between global and local knowledge mastery states. To bridge these gaps, this paper introduces a novel Learning State Enhanced Co-attention Model (LSEKT) for knowledge tracing. In terms of methodology, we contend that a student’s recent answering behavior is intricately tied to implicit learning states. Consequently, we devise a learning state extraction network to capture the student’s current learning state. Furthermore, to construct a more robust and interdependent representation of both global and local knowledge mastery states, we integrate a co-attention network. This network enhances the attention paid to pertinent knowledge points across both global and local scales, thereby adeptly capturing the underlying connections between global and local interaction sequences. Concurrently, we incorporate contrastive learning as an auxiliary task within our model to bolster its predictive prowess. Ultimately, we evaluated our approach through extensive experiments on four widely used datasets. The experimental outcomes underscore the remarkable performance of our model across diverse evaluation metrics, emphasizing the effectiveness of our proposed LSEKT model.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761742","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":"StealthMask: Highly stealthy adversarial attack on face recognition system","authors":"Jian-Xun Mi,&nbsp;Mingxuan Chen,&nbsp;Tao Chen,&nbsp;Xiao Cheng","doi":"10.1007/s10489-025-06511-4","DOIUrl":"10.1007/s10489-025-06511-4","url":null,"abstract":"<div><p>Convolutional Neural Networks (CNNs) based on deep learning algorithms are widely used in real-world scenarios. However, these networks are vulnerable to adversarial examples-maliciously crafted inputs that can cause the model to make incorrect predictions. The existence of adversarial examples presents a significant challenge to the field of deep learning, with profound implications for various aspects of our lives. In face recognition technology, adversarial examples pose a substantial security risk. In this paper, we propose a novel method for generating adversarial patches designed to be worn as masks. The perturbed mask is crafted to deceive face recognition models, thereby highlighting the security vulnerabilities inherent in this technology. Our experimental results demonstrate that the mask generated by the proposed method effectively misleads the face recognition system, achieving high attack success rates while maintaining necessary stealthiness and transferability. Moreover, our method successfully attacks commercial face recognition systems and real-world access control systems, exposing the vulnerabilities of existing face recognition technologies in security-critical applications. Notably, compared to traditional methods, our proposed method emphasizes the stealthiness of the adversarial mask more than traditional methods. To account for physical-world factors, such as distortion, rotation, and deformations, we integrate a specifically designed loss function, thereby enhancing the method’s stability and reliability in practical scenarios.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769832","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":"Unsupervised learning with physics informed graph networks for partial differential equations","authors":"Lin Lu,&nbsp;Yiye Zou,&nbsp;Jingyu Wang,&nbsp;Shufan Zou,&nbsp;Laiping Zhang,&nbsp;Xiaogang Deng","doi":"10.1007/s10489-025-06479-1","DOIUrl":"10.1007/s10489-025-06479-1","url":null,"abstract":"<div><p>Natural physical phenomena are commonly expressed using partial differential equations (PDEs), in domains such as fluid dynamics, electromagnetism, and atmospheric science. These equations typically require numerical solutions under given boundary conditions. There is a burgeoning interest in the exploration of neural network methodologies for solving PDEs, mainly based on automatic differentiation methods to learn the PDE-solving process, which means that the model needs to be retrained when the boundary conditions of PDE are changed. However, automatic differentiation requires substantial memory resources to facilitate the training regimen. Moreover, a learning objective that is tailored to the solution process often lacks the flexibility to extend to boundary conditions; thereby limiting the solution’s overall precision. The method proposed in this paper introduces a graph neural network approach, embedded with physical information, mainly for solving Poisson’s equation. An approach is introduced that reduces memory usage and enhances training efficiency through an unsupervised learning methodology based on numerical differentiation. Concurrently, by integrating boundary conditions directly into the neural network as supplementary physical information, this approach ensures that a singular model is capable of solving PDEs across a variety of boundary conditions. To address the challenges posed by more complex network inputs, the introduction of graph residual connections serves as a strategic measure to prevent network overfitting and to elevate the accuracy of the solutions provided. Experimental findings reveal that, despite having 30 times more training parameters than the Physics-Informed Neural Networks (PINN) model, the proposed model consumes 2.2% less memory than PINN. Additionally, generalization in boundary conditions has been achieved to a certain extent. This enables the model to solve partial differential equations with different boundary conditions, a capability that PINN currently lacks. To validate the solving capability of the proposed method, it has been applied to the model equation, the Sod shock tube problem, and the two-dimensional inviscid airfoil problem. In terms of the solution accuracy of the model equations, the proposed method outperforms PINN by 30% to four orders of magnitude. Compared to the traditional numerical method, the Finite Element Method (FEM), the proposed method also shows an order of magnitude improvement. Additionally, when compared to the improved version of PINN, TSONN, our method demonstrates certain advantages. The forward problem of the Sod shock tube, which PINN is currently unable to solve, is successfully handled by the proposed method. For the airfoil problem, the results are comparable to those of PINN.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761711","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":"Topology-preserving and structure-aware (hyper)graph contrastive learning for node classification","authors":"Minhao Zou,&nbsp;Zhongxue Gan,&nbsp;Yutong Wang,&nbsp;Junheng Zhang,&nbsp;Chun Guan,&nbsp;Siyang Leng","doi":"10.1007/s10489-025-06491-5","DOIUrl":"10.1007/s10489-025-06491-5","url":null,"abstract":"<div><p>Recently, graph contrastive learning (GCL) has attracted considerable attention, establishing a new paradigm for learning graph representations in the absence of human annotations. While notable advancements have been made, simultaneous consideration of both graphs and hypergraphs remains rare. This limitation arises because graphs and hypergraphs encode connectivity differently, making it challenging to develop a unified structure augmentation strategy. Conventional structure augmentation methods like adding or removing edges risk imperiling intrinsic topological traits and introducing adverse distortions such as disconnected subgraphs or isolated nodes. In this work, we propose a framework of contrastive learning on graphs and hypergraphs, named as UniGCL, to address these challenges by leveraging a unified adjacency representation that enables simultaneous modeling of pairwise and higher-order relationships. In particular, two structure augmentation methods are developed to perturb graph structure weights instead of altering connectivity, thereby preserving both graph and hypergraph topology while generating diverse augmented views. Furthermore, a structure-aware contrastive loss is proposed, which incorporates gradient perturbation techniques to enhance the model’s ability to capture fine-grained structural dependencies in (hyper)graphs. Extensive experiments are conducted on six real-world graph datasets and nine representative hypergraph datasets to evaluate the performance of the proposed framework. The results demonstrate that UniGCL achieves superior node classification performance compared to the advanced graph and hypergraph contrastive learning methods, across datasets with different homophilic extents and limited annotations. Additionally, ablation studies validate the effectiveness of our structure-preserving augmentations and structure-aware contrastive loss in enhancing performance.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761741","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":"Intra-frame scan-free video state spaces model for video moment retrieval","authors":"Fengzhen Yu,&nbsp;Xiaodong Gu","doi":"10.1007/s10489-025-06517-y","DOIUrl":"10.1007/s10489-025-06517-y","url":null,"abstract":"<div><p>With the increasing complexity of video moment retrieval tasks, effectively handling temporal and spatial information in video data has become a central challenge. This paper proposes a novel Intra-frame Scan-free Video State Spaces Model to address the spatiotemporal modeling problem in video moment retrieval. The model eliminates the dependency on the scanning order of intra-frame patches, overcoming the dual temporal limitations of frame order and within-frame patch sequence, which enhances the flexibility and efficiency of video understanding. To better model temporal information, we introduce the concept of video moment boundaries and propose the Weighted Relative Center Difference Loss, which ensures that the predicted center regions are closer to the ground truth, thereby improving retrieval accuracy. Extensive experiments on three public video datasets (ActivityNet Captions, TACoS, and Charades-STA) show that the model achieves superior or near-optimal performance across multiple metrics. The ablation study compares the performance loss when removing different components, the effect of different scanning methods on performance and inference throughput, and the effect of hyperparameters such as the number of SSM layers and the weighted relative centre difference loss threshold on retrieval performance. These results validate the effectiveness and robustness of our approach for video moment retrieval.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769822","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":"Multiple disease diagnoses using heterogeneous EHR curated knowledge graph and machine learning models","authors":"Shivani Dhiman,&nbsp;Anjali Thukral,&nbsp;Punam Bedi","doi":"10.1007/s10489-024-05952-7","DOIUrl":"10.1007/s10489-024-05952-7","url":null,"abstract":"<div><p>Artificial Intelligence (AI) can play a significant role by assisting healthcare professionals in disease diagnosis, which is a critical step towards a patient’s treatment. Most of the research work in disease diagnosis systems predicts the presence or absence of a given single disease in a patient. However, there are only a few studies on multiple disease diagnoses, i.e., on detecting the presence of more than one disease at the same time. In this paper, we propose a framework for diagnosing multiple diseases using Knowledge Graph (KG), Knowledge embeddings and Machine Learning (ML). KG is created to semantically organize heterogeneous clinical details extracted from Electronic Health Records (EHRs). Additionally, we present a detailed comparison and analysis of three disease diagnosis systems, Single Disease Single Diagnosis (SDSD), Multiple Disease Single Diagnosis (MDSD), and Multiple Disease Multiple Diagnosis (MDMD) using the MIMIC-III dataset on Chronic Heart Failure (CHF), Acute Respiratory Failure (ARF) and Acute Kidney Failure (AKF) diseases. The above disease diagnosis systems have been implemented and analysed with different ML algorithms, such as Logistic Regression (LR), Random Forest (RF), Naïve Bayes (NB), and Support Vector Machine (SVM). Besides, detecting the probability of having multiple diseases at a time, the MDMD shows comparable results in comparison to SDSD and MDSD. This is being evaluated by using the Area Under Receiver Operating Characteristic (AUROC) and the Area Under Precision-Recall Curve (AUPRC) metrics. The MDMD system based on the proposed framework for multiple disease diagnosis predicts CHF, ARF and AKF in 91%, 74% and 79% of positive cases, respectively.</p><h3>Graphical Abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143769824","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":"A modified SimRank++ approach for searching crash simulation data","authors":"Anahita Pakiman,&nbsp;Jochen Garcke,&nbsp;Axel Schumacher","doi":"10.1007/s10489-024-05945-6","DOIUrl":"10.1007/s10489-024-05945-6","url":null,"abstract":"<div><p>Data searchability has been utilized for decades and is now a crucial ingredient of data reuse. However, data searchability in industrial engineering is essentially still at the level of individual text documents, while for finite element (FE) simulations no content-based relations between FE simulations exist so far. Additionally, the growth of data warehouses with the increase of computational power leaves companies with a vast amount of engineering data that is rarely reused. Search techniques for FE data, which are in particular aware of the engineering problem context, is a new research topic. We introduce the prediction of similarities between simulations using graph algorithms, which for example allows the identification of outliers or ranks simulations according to their similarities. With that, we address searchability for FE-based crash simulations in the automotive industry. Here, we use SimRank-based methods to predict the similarity of crash simulations using unweighted and weighted bipartite graphs. Motivated by requirements from the engineering application, we introduce SimRankTarget++ an alternative formulation of SimRank++ that performs better for FE simulations. To show the generality of the graph approach, we compare component-based similarities with part-based ones. For that, we introduce a method for automatically detecting components in the vehicle. We use a car sub-model to illustrate the similarity ansatz and present results on data from real-life development stages of an automotive company.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 7","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-024-05945-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143761696","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}