Li Wang , Lucong Zhang , Haotian Wu , Teng Zhang , Ke Qiu , Tianyu Chen , Hongwu Qin
{"title":"Multisource-domain regression transfer learning framework for predicting student academic performance considering balanced similarity","authors":"Li Wang , Lucong Zhang , Haotian Wu , Teng Zhang , Ke Qiu , Tianyu Chen , Hongwu Qin","doi":"10.1016/j.engappai.2025.111202","DOIUrl":"10.1016/j.engappai.2025.111202","url":null,"abstract":"<div><div>The increasing integration of information technology and artificial intelligence has extensively implemented computer-aided intelligent education systems in higher education. A critical task within these systems is student performance prediction, which forecasts future academic outcomes by analyzing data such as historical grades, learning behaviors, and classroom participation. This enables early intervention and personalized teaching based on scientific evidence. However, most existing methods rely on traditional machine learning techniques, which can hardly address issues such as domain distribution discrepancies and data imbalance effectively. To overcome these challenges, we propose a multisource-domain transfer learning regression framework that integrates domain selection, hybrid feature extraction, and dynamic joint distribution adaptation techniques. Specifically, the framework first selects appropriate source domains on the basis of preset thresholds via cross-validation. Thereafter, a hybrid feature extractor is used to derive (i) common features from the target and selected source domains and (ii) domain-specific features from the target domain. Finally, a dynamic adaptive factor is introduced to balance differences between the marginal and conditional distributions. Experimental results indicate that the proposed framework significantly reduces the root mean square error with an average prediction improvement of 21.05 %, compared with baseline methods and other advanced approaches.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111202"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169112","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":"Hybrid aero-engine performance modeling to enable real-time capability using physics-based analysis and machine learning","authors":"Sangjo Kim","doi":"10.1016/j.engappai.2025.111288","DOIUrl":"10.1016/j.engappai.2025.111288","url":null,"abstract":"<div><div>The ability to achieve rapid and efficient computation is critical for real-time analysis or onboard implementation of aero engine performance models. This study presents a hybrid aero engine performance modeling approach that combines the accuracy of component map-based zero-dimensional cycle analysis with the speed of machine learning. Traditional models using iterative solvers like Newton-Raphson are computationally intensive and prone to convergence issues. To address this, a feedforward neural network was trained on data from diverse steady-state and dynamic conditions to replace the iterative solver. The machine learning -based solver reduced execution time by approximately 90 % while maintaining predictive accuracy. Applied to the F404 engine, the proposed method showed high agreement with conventional models under steady-state conditions and acceptable performance under dynamic scenarios. The distinct advantages of this integrated approach include significant computational savings, enhanced adaptability to operational changes, and improved stability, supporting real-time, onboard performance analysis and digital twin applications.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111288"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169265","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-regularized tensor-based framework for identifying hard landings","authors":"Chenyang Chang , Yu An , Xi Zhang","doi":"10.1016/j.engappai.2025.111178","DOIUrl":"10.1016/j.engappai.2025.111178","url":null,"abstract":"<div><div>Hard landings are a significant concern in civil aviation, often resulting in aircraft structural damage, financial losses, and compromised passenger safety. Automating the detection of such incidents faces challenges due to the complexities of Quick Access Recorder (QAR) data, which exhibit multi-channel interdependencies and temporal dynamics. Furthermore, environmental factors tied to flight-variant data, such as the geographic attributes of landing airports, can influence the occurrence of hard landings, yet these factors are often neglected in existing methodologies. This omission limits the practical utility of current approaches for enhancing safety in civil aviation. To address these challenges, we propose a multi-regularized tensor-based framework that models QAR data as a high-order tensor and applies tensor decomposition to extract latent patterns that characterize hard landing scenarios. The model incorporates tailored regularization terms to address both temporal correlations and inter-channel couplings across aircraft systems. To enable efficient computation, we develop a customized Block Coordinate Descent (BCD) algorithm, designed for efficient processing with high-dimensional factor matrices. The effectiveness of the proposed framework is validated using real-world civil aviation data from China, demonstrating superior performance in identifying hard landings.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111178"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169341","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}
Huajian Zhang, Shuhai Jia, Bo Wen, Xing Zhou, Zihan Lin, Longning Wang, Mengyu Han
{"title":"Utilizing convolutional neural networks for calculating full-field stress components and directions in photoelasticity","authors":"Huajian Zhang, Shuhai Jia, Bo Wen, Xing Zhou, Zihan Lin, Longning Wang, Mengyu Han","doi":"10.1016/j.engappai.2025.111173","DOIUrl":"10.1016/j.engappai.2025.111173","url":null,"abstract":"<div><div>Photoelasticity is a crucial experimental technique extensively used in various engineering and scientific domains. The integration of convolutional neural networks can substantially enhance the efficiency and capability of photoelasticity in resolving full-field stress. Nevertheless, existing neural network–based methods in photoelasticity are limited to computing only the difference between the principal stresses rather than determining the individual principal stress components (i.e., the absolute values of the first and second principal stresses) and the principal stress direction (isoclinic parameter). Directly solving for the principal stress components is of greater significance in many practical problems, and the principal stress direction is essential for evaluating material failure and optimizing designs. In this paper, a convolutional neural network is proposed for the first time that can directly and simultaneously determine the full-field first principal stress, second principal stress, and the principal stress direction. A dataset generation method was developed to train this network, producing a novel high-quality dataset containing 41,000 raw samples, without data augmentation. The proposed network exhibits high accuracy and strong generalization across synthetic and experimental validation sets. On the synthesized dataset, the structural similarity exceeds 0.98, and the mean squared error is below 0.45, with similarly satisfactory results on the experimental validation sets. This network establishes a direct mapping between photoelastic images and full-field stress components and directions, thereby enhancing the efficiency and potential applications of photoelasticity. The proposed dataset generation method may also offer valuable insights for advancing deep learning in photoelasticity.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111173"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169185","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":"Transformers in pathological image analysis: A survey","authors":"Liangliang Liu, Zhihong Liu, Jinpu Xie, Hongbo Qiao, Jing Chang","doi":"10.1016/j.engappai.2025.111114","DOIUrl":"10.1016/j.engappai.2025.111114","url":null,"abstract":"<div><div>With the advancements of artificial intelligence, deep learning has emerged as the predominant approach in computational pathology. It is dedicated to automatically analyzing the intricate phenotype information embedded in various pathological images, with the goal of delivering more precise diagnoses, prognoses, and treatment recommendations for cancer patients. As the latest breakthrough in deep learning technology, Transformers are gaining traction in the realm of pathological image analysis by harnessing self-attention mechanisms to capture global information. Consequently, this study presents a comprehensive review of state-of-the-art models leveraging Transformers, applied across tasks such as classification, segmentation, and survival analysis in pathological image analysis. Initially, we delineate the concept and key components of Transformers, followed by a survey of their recent applications in pathology. These applications encompass pathological image classification, segmentation, lesion detection and localization, as well as the utilization of specific Transformer architectures for patient survival analysis. Subsequently, we delve into the challenges encountered in employing Transformers for pathological image analysis and speculate on future developmental trajectories. Our aim is to furnish readers with an exhaustive roadmap to deepen their comprehension of Transformer applications in pathology, thereby fostering the advancement of more sophisticated technologies and enabling more precise diagnoses and treatment strategies for cancer patients.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111114"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169077","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":"CTPR: Contrastive transition predictive representation for reinforcement learning","authors":"Hao Sun , Changpeng Wang","doi":"10.1016/j.engappai.2025.111124","DOIUrl":"10.1016/j.engappai.2025.111124","url":null,"abstract":"<div><div>Learning policies from high-dimensional observations is a challenging problem for pixel-based reinforcement learning. Most existing pixel-based reinforcement learning methods struggle with the inefficiency of extracting meaningful state representations from raw pixel data, lacking temporal correlation and resulting in suboptimal performance. To this end, we propose a innovative method named contrastive transition predictive representation for reinforcement learning (CTPR), which utilizes contrastive learning and a transition model to efficiently extract high-level state representations from raw pixels for sample-efficient reinforcement learning. In the reinforcement learning component, we perform policy control based on the learned contrastive representations. We have evaluated the effectiveness of the proposed method by conducting numerous experiments on DeepMind Control, and the results show that our method has achieve significant improvements over the state-of-the-art methods.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111124"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169205","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":"Role of noise elimination algorithms in speech processing applications: A comprehensive research and some experimental results","authors":"Nagaraja B.G. , Thimmaraja Yadava G. , Raghudathesh G.P. , Jayanna H.S.","doi":"10.1016/j.engappai.2025.111116","DOIUrl":"10.1016/j.engappai.2025.111116","url":null,"abstract":"<div><div>The performance of speech-based systems is severely degraded due to the presence of background noise in real-world environments. Effective noise elimination algorithms are essential for enhancing speech quality and improving the performance of speech processing applications, such as voice activity detection (VAD) and speech encoding. Various speech enhancement techniques have been proposed to tackle this, and in this context, choosing an appropriate enhancement technique for improving speech quality and intelligibility is an important consideration. This paper presents a concise experimental review of different noise elimination techniques using objective and subjective metrics. The experiments are conducted on the noisy speech corpus (NOIZEUS) across different noise types and signal-to-noise ratio (SNR) levels. Comparative results indicate that the soft mask estimator with a <em>priori</em> SNR uncertainty (SMPR) is considerably more useful in enhancing speech quality. Furthermore, we analyze the SMPR performance in enhancing speech quality under various noise conditions, specifically focusing on their impact on speech encoding and VAD applications. Our results reveal that integrating the SMPR enhancement module into linear predictive coding (LPC)-based speech encoding system significantly improves speech quality. Additionally, the application of SMPR in VAD systems demonstrates notable improvements, enhancing the accuracy of speech detection.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111116"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169189","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":"Fire resistance rating prediction of timber-to-steel connections and design optimization informed by explainable machine learning","authors":"Tongchen Han , Zhidong Zhang , Weiwei Wu","doi":"10.1016/j.engappai.2025.111127","DOIUrl":"10.1016/j.engappai.2025.111127","url":null,"abstract":"<div><div>Timber as a construction material is experiencing its renaissance, while fire safety is a critical factor for timber-based building design. Currently, the fire resistance rating of wood-steel-wood (WSW) connections is evaluated using empirical equations derived from experimental results. However, these equations consider a limited set of parameters and lack interpretability. This paper developed an explainable machine learning (ML) model considering comprehensive parameters related to connection’s configuration, based on 140 experimental and experimental-validated numerical data. The performances of various machine learning models are evaluated in terms of predicting the fire resistance rating of connections after hyperparameter tuning. The eXtreme Gradient Boosting (XGBoost) model outperforms other ML models (<span><math><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup><mo>=</mo><mn>0</mn><mo>.</mo><mn>93</mn></mrow></math></span>) and empirical equations. The local sensitivity analysis (LSA), global sensitivity analysis (GSA), and SHapley Additive exPlanations (SHAP) analysis are conducted based on the XGBoost model to investigate the contributions of nine parameters to the fire resistance rating. Both sensitivity analysis and SHAP analysis identify timber thickness and load ratio as the primary factors influencing fire resistance. Finally, the calibrated XGBoost model is incorporated into a non-dominated sorting genetic algorithm (NSGA-II) to optimize the design, aiming to minimize the self-weight of the connection while maximizing the fire resistance rating and load-carrying capacity of the connection subjected to constraints on limited dimensions.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111127"},"PeriodicalIF":7.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144169192","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":"Graph Neural Network-based Deep Reinforcement Learning algorithm for Virtual Network Function forwarding graph embedding in Space–Air–Ground Integrated Network","authors":"Liang Liu, Tengxiang Jing, Siyuan Tan, Yujie Zhang, Yejun He, Chuan Xu","doi":"10.1016/j.engappai.2025.111083","DOIUrl":"10.1016/j.engappai.2025.111083","url":null,"abstract":"<div><div>Space–Air–Ground Integrated Network (SAGIN) can provide seamless three-dimensional coverage and enhanced flexibility, and it has been recognized as the core architecture of future 6G networks. Software Defined Network (SDN) and Network Function Virtualization (NFV) are two enabling technologies of SAGIN, which can sequentially arrange multiple Virtual Network Functions (VNFs) into VNF Forwarding Graph (VNF-FG) to provide users with scalable and parallel network services. However, SAGIN exhibits significant dynamism and heterogeneity, and VNFs may be deployed in multiple different heterogeneous locations, which brings great challenges to the efficient embedding of VNF-FGs required for user service request flows. In this paper, we study the VNF-FG embedding problem by jointly considering the structural constraints, node and link resource constraints, and End-to-End (E2E) delay constraint of VNF-FG in SDN/NFV-enabled SAGIN. Specifically, we first design a three-layer SDN/NFV-enabled SAGIN architecture consisting of a global controller and distributed intra-domain SDN controllers. Then, we define the Delay and Cost Efficient Dynamic VNF-FG Embedding Problem (DCE-DVEP) and formulate it as an Integer Linear Programming (ILP) with the objective of minimizing the weighted sum of E2E delay and embedding cost of all VNF-FGs. Finally, a Graph Neural Network-based Deep Reinforcement learning Embedding (GNN-DRE) algorithm is proposed to solve the DCE-DVEP, which can more accurately capture the rich feature information from both SAGIN and VNF-FG by specifically integrating different GNN models and adopts Deep Reinforcement Learning (DRL) to make effective embedding decisions. The simulation results demonstrate that, compared with other baseline algorithms, the GNN-DRE can reduce E2E delay and embedding cost by about 8% and 13%, respectively.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"156 ","pages":"Article 111083"},"PeriodicalIF":7.5,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144147014","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}