计算机科学最新文献

英文中文

IF： -

Expert-level policy style measurement via knowledge distillation with large language model collaboration

IF 7.4 1区管理学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Processing & Management

Pub Date : 2025-02-22 DOI: 10.1016/j.ipm.2025.104090

Yujie Zhang , Biao Huang , Weikang Yuan , Zhuoren Jiang , Longsheng Peng , Shuai Chen , Jie-Sheng Tan-Soo

Policy style is a crucial concept in policy science that reflects persistent patterns in the policy process across different governance settings. Despite its importance, policy style measurement faces issues of complexity, subjectivity, data sparseness, and computational cost. To overcome these obstacles, we propose KOALA, a novel KnOwledge distillation framework based on large lAnguage modeL collAboration. It transforms the weak scoring abilities of LLMs into a pairwise ranking problem, employs a small set of expert-annotated samples for non-parametric learning, and utilizes knowledge distillation to transfer insights from LLMs to a smaller, more efficient model. The framework incorporates multiple LLM-based agents (Prompter, Ranker, and Analyst) collaborating to comprehend complex measurement standards and self-explain policy style definitions. We validate KOALA on 4,572 Chinese government work reports (1954–2019) from central, provincial, and municipal levels, with a focus on the imposition dimension of policy style. Extensive experiments demonstrate KOALA’s effectiveness in measuring the intensity of policy style, highlighting its superiority over state-of-the-art methods. While GPT-4 achieves only 66% accuracy in pairwise ranking of policy styles, KOALA, despite being based on GPT-3.5, achieves a remarkable 85% accuracy, highlighting significant performance improvement. This framework offers a transferable approach for quantifying complex social science concepts in textual data, bridging computational techniques with social science research.

{"title":"Expert-level policy style measurement via knowledge distillation with large language model collaboration","authors":"Yujie Zhang , Biao Huang , Weikang Yuan , Zhuoren Jiang , Longsheng Peng , Shuai Chen , Jie-Sheng Tan-Soo","doi":"10.1016/j.ipm.2025.104090","DOIUrl":"10.1016/j.ipm.2025.104090","url":null,"abstract":"<div><div>Policy style is a crucial concept in policy science that reflects persistent patterns in the policy process across different governance settings. Despite its importance, policy style measurement faces issues of complexity, subjectivity, data sparseness, and computational cost. To overcome these obstacles, we propose <strong>KOALA</strong>, a novel <strong><u>K</u></strong>n<strong><u>O</u></strong>wledge distillation framework based on large l<strong><u>A</u></strong>nguage mode<strong><u>L</u></strong> coll<strong><u>A</u></strong>boration. It transforms the weak scoring abilities of LLMs into a pairwise ranking problem, employs a small set of expert-annotated samples for non-parametric learning, and utilizes knowledge distillation to transfer insights from LLMs to a smaller, more efficient model. The framework incorporates multiple LLM-based agents (Prompter, Ranker, and Analyst) collaborating to comprehend complex measurement standards and self-explain policy style definitions. We validate KOALA on 4,572 Chinese government work reports (1954–2019) from central, provincial, and municipal levels, with a focus on the imposition dimension of policy style. Extensive experiments demonstrate KOALA’s effectiveness in measuring the intensity of policy style, highlighting its superiority over state-of-the-art methods. While GPT-4 achieves only 66% accuracy in pairwise ranking of policy styles, KOALA, despite being based on GPT-3.5, achieves a remarkable 85% accuracy, highlighting significant performance improvement. This framework offers a transferable approach for quantifying complex social science concepts in textual data, bridging computational techniques with social science research.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"62 4","pages":"Article 104090"},"PeriodicalIF":7.4,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471177","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Meta-machine learning framework for robust short-term solar power prediction across different climatic zones

IF 7.5 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Engineering Applications of Artificial Intelligence

Pub Date : 2025-02-22 DOI: 10.1016/j.engappai.2025.110295

Amit Rai , Ashish Shrivastava , Kartick C. Jana , Jay Liu , Kulwant Singh , N.S. Jayalakshmi , Amit Agrawal

The global energy landscape is increasingly dominated by solar power installations, driven by the sun’s position as Earth’s most abundant and sustainable energy resource. However, the intermittent nature of solar radiation, influenced by both astronomical cycles and meteorological conditions, creates significant challenges for reliable power generation and grid integration. To address the issue of uncertainty, this study proposes a robust and improved capacity machine learning framework with enhanced hypothesis functional space. The proposed model improves the capacity of an individual model by combining the hypothesis functions of individual machine learning models, increasing the representational capacity and hence the model’s generalization. Moreover, a non-linear second stage is stacked to increase the depth of the proposed model, which utilizes meta-data of first stage to further improve the forecasting accuracy. Furthermore, the proposed model is validated on four different climatic zones of the world for solar power forecasting. The proposed model achieves an average improvement of 66.7% in mean absolute error across all locations compared to the next best performing algorithm, with particularly strong performance in arid zones. Statistical validation through Cook’s distance analysis also confirms the model’s reliability with an average of 8.64% influential points across all locations.

{"title":"Meta-machine learning framework for robust short-term solar power prediction across different climatic zones","authors":"Amit Rai , Ashish Shrivastava , Kartick C. Jana , Jay Liu , Kulwant Singh , N.S. Jayalakshmi , Amit Agrawal","doi":"10.1016/j.engappai.2025.110295","DOIUrl":"10.1016/j.engappai.2025.110295","url":null,"abstract":"<div><div>The global energy landscape is increasingly dominated by solar power installations, driven by the sun’s position as Earth’s most abundant and sustainable energy resource. However, the intermittent nature of solar radiation, influenced by both astronomical cycles and meteorological conditions, creates significant challenges for reliable power generation and grid integration. To address the issue of uncertainty, this study proposes a robust and improved capacity machine learning framework with enhanced hypothesis functional space. The proposed model improves the capacity of an individual model by combining the hypothesis functions of individual machine learning models, increasing the representational capacity and hence the model’s generalization. Moreover, a non-linear second stage is stacked to increase the depth of the proposed model, which utilizes meta-data of first stage to further improve the forecasting accuracy. Furthermore, the proposed model is validated on four different climatic zones of the world for solar power forecasting. The proposed model achieves an average improvement of 66.7% in mean absolute error across all locations compared to the next best performing algorithm, with particularly strong performance in arid zones. Statistical validation through Cook’s distance analysis also confirms the model’s reliability with an average of 8.64% influential points across all locations.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"147 ","pages":"Article 110295"},"PeriodicalIF":7.5,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471368","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}

引用次数: 0

Stabilization of unstable impulsive systems via stochastic discrete-time feedback control with Lévy noise

IF 3.7 2区计算机科学 Q2 AUTOMATION & CONTROL SYSTEMS

Nonlinear Analysis-Hybrid Systems

Pub Date : 2025-02-22 DOI: 10.1016/j.nahs.2025.101585

Mengmeng Zhang, Quanxin Zhu

It is universally acknowledged that noise can stabilize systems. Given an unstable impulsive system, can it be stabilized by Lévy noise? In this article, we mainly discuss the stochastic stabilization of unstable impulsive systems via Lévy noise. First, the conditions for

p

-moment exponential stability of unstable impulsive systems under continuous-time feedback control with Lévy noise are derived using the Lyapunov function method. Second, the almost sure exponentially stable conditions for the unstable impulsive systems under discrete-time feedback control with Lévy noise are derived through the comparison method. Additionally, several new stochastic stabilization strategies are proposed to design stochastic discrete-time feedback control for deriving almost sure exponentially stable results through stochastic analysis. Finally, two examples and their relevant simulation figures are given to check the validity of the results.

众所周知，噪声可以稳定系统。给定一个不稳定的脉冲系统，它能否被勒维噪声稳定呢？本文主要讨论通过莱维噪声实现不稳定脉冲系统的随机稳定。首先，利用莱维噪声的连续时间反馈控制，推导了不稳定脉冲系统的 p时刻指数稳定条件。其次，通过比较法推导了具有 Lévy 噪声的离散时间反馈控制下不稳定脉冲系统的几乎确定的指数稳定条件。此外，还提出了几种新的随机稳定策略，用于设计随机离散时间反馈控制，通过随机分析得出几乎确定的指数稳定结果。最后，给出了两个实例及其相关仿真数据，以检验结果的正确性。

引用次数: 0

Enhancement design of eleven-level cascaded h-bridge motor driver application

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-02-22 DOI: 10.1016/j.compeleceng.2025.110179

Adil Adam , Firat Kacar , Nikos Mastorakis

This study focuses on the design and implementation of an innovative eleven-level cascaded H-Bridge motor drive (11L-CHBMD) controlled by a three-phase step-sine pulse width modulation (SSPWM) technique. A novel mathematical model was developed by converting control equations into matrix format, facilitating precise simulation and practical realization of the system. MATLAB Simulink was employed for the simulation, while the STM32F429ZGT6 microcontroller and power MOSFETs were used for hardware implementation. The proposed system ensures a regulated high-voltage, variable-current output and achieves harmonic distortion levels below 5 %, in compliance with IEEE-519 standards. Experimental results showed the motor driver 11L-CHBMD's high capability to drive three-phase induction motors efficiently, offering superior performance compared to conventional topologies. The SSPWM method reduced total harmonic distortion (THD) while maintaining system stability under ohmic, inductive, and unbalanced load conditions. Fuzzy and PID controllers enabled precise torque, speed, and current regulation while stabilising faster. The 11L-CHBMD proposed circuit, developed using commonly available components, achieves a cost reduction of approximately 90 % compared to market-available designs, making it suitable for industrial, renewable energy and different applications. Its modular design supports scalability and offers potential for driving motors in hazardous environments or remote areas using solar energy. With its adaptability and efficiency, the proposed 11L-CHBMD stands as a compelling alternative to traditional power inverters.

{"title":"Enhancement design of eleven-level cascaded h-bridge motor driver application","authors":"Adil Adam , Firat Kacar , Nikos Mastorakis","doi":"10.1016/j.compeleceng.2025.110179","DOIUrl":"10.1016/j.compeleceng.2025.110179","url":null,"abstract":"<div><div>This study focuses on the design and implementation of an innovative eleven-level cascaded H-Bridge motor drive (11L-CHBMD) controlled by a three-phase step-sine pulse width modulation (SSPWM) technique. A novel mathematical model was developed by converting control equations into matrix format, facilitating precise simulation and practical realization of the system. MATLAB Simulink was employed for the simulation, while the STM32F429ZGT6 microcontroller and power MOSFETs were used for hardware implementation. The proposed system ensures a regulated high-voltage, variable-current output and achieves harmonic distortion levels below 5 %, in compliance with IEEE-519 standards. Experimental results showed the motor driver 11L-CHBMD's high capability to drive three-phase induction motors efficiently, offering superior performance compared to conventional topologies. The SSPWM method reduced total harmonic distortion (THD) while maintaining system stability under ohmic, inductive, and unbalanced load conditions. Fuzzy and PID controllers enabled precise torque, speed, and current regulation while stabilising faster. The 11L-CHBMD proposed circuit, developed using commonly available components, achieves a cost reduction of approximately 90 % compared to market-available designs, making it suitable for industrial, renewable energy and different applications. Its modular design supports scalability and offers potential for driving motors in hazardous environments or remote areas using solar energy. With its adaptability and efficiency, the proposed 11L-CHBMD stands as a compelling alternative to traditional power inverters.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110179"},"PeriodicalIF":4.0,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hybrid deep-learning prediction model based on kernel multi-granularity fuzzy rough sets and its application in the diagnosis and treatment of chronic kidney disease

IF 7.5 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Engineering Applications of Artificial Intelligence

Pub Date : 2025-02-22 DOI: 10.1016/j.engappai.2025.110297

Jiqian Liu , Bingzhen Sun , Jin Ye , Xixuan Zhao , Xiaoli Chu

In uncertain decision-making scenarios, quantitative scientific prediction models and methods can provide valuable support for making scientific decisions. However, the characteristics of hybrid attribute information may lead to a series of issues. These include difficulties in comparing and comprehensively evaluating different types of attributes, nonlinear relationships between attributes, and a lack of effective decision-support methods. To overcome these issues, this study introduces a kernel function to abstract the similarity of different attribute types and proposes a model called kernel multi-granularity fuzzy rough sets (KMGFRS). The KMGFRS model facilitates a thorough exploration and analysis of the uncertainties associated with decision objects. Additionally, an attribute reduction method based on KMGFRS is discussed to address redundant attributes in hybrid information systems. This method eliminates attributes that have a minimal influence on the decision results, simplifies the decision process, and enhances its effectiveness. This study integrates the KMGFRS and hybrid deep learning concepts to propose a novel prediction method aimed at enhancing accuracy and robustness. From the perspective of hybrid attribute information, this method can more accurately predict the unknown attributes of decision objects, thereby providing robust support for disease prediction in medical diagnostics and therapeutic decision-making. The experimental results indicated that the constructed model effectively handled uncertain decision-making scenarios involving hybrid attributes and fuzzy decision objects. It provides accurate and reliable decision support for chronic kidney disease (CKD), significantly enhancing the predictive accuracy of CKD types.

{"title":"Hybrid deep-learning prediction model based on kernel multi-granularity fuzzy rough sets and its application in the diagnosis and treatment of chronic kidney disease","authors":"Jiqian Liu , Bingzhen Sun , Jin Ye , Xixuan Zhao , Xiaoli Chu","doi":"10.1016/j.engappai.2025.110297","DOIUrl":"10.1016/j.engappai.2025.110297","url":null,"abstract":"<div><div>In uncertain decision-making scenarios, quantitative scientific prediction models and methods can provide valuable support for making scientific decisions. However, the characteristics of hybrid attribute information may lead to a series of issues. These include difficulties in comparing and comprehensively evaluating different types of attributes, nonlinear relationships between attributes, and a lack of effective decision-support methods. To overcome these issues, this study introduces a kernel function to abstract the similarity of different attribute types and proposes a model called kernel multi-granularity fuzzy rough sets (KMGFRS). The KMGFRS model facilitates a thorough exploration and analysis of the uncertainties associated with decision objects. Additionally, an attribute reduction method based on KMGFRS is discussed to address redundant attributes in hybrid information systems. This method eliminates attributes that have a minimal influence on the decision results, simplifies the decision process, and enhances its effectiveness. This study integrates the KMGFRS and hybrid deep learning concepts to propose a novel prediction method aimed at enhancing accuracy and robustness. From the perspective of hybrid attribute information, this method can more accurately predict the unknown attributes of decision objects, thereby providing robust support for disease prediction in medical diagnostics and therapeutic decision-making. The experimental results indicated that the constructed model effectively handled uncertain decision-making scenarios involving hybrid attributes and fuzzy decision objects. It provides accurate and reliable decision support for chronic kidney disease (CKD), significantly enhancing the predictive accuracy of CKD types.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"147 ","pages":"Article 110297"},"PeriodicalIF":7.5,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471369","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}

引用次数: 0

Self-supervised heterogeneous graph neural network based on deep and broad neighborhood encoding

IF 3.4 2区计算机科学 Q2 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Applied Intelligence

Pub Date : 2025-02-22 DOI: 10.1007/s10489-025-06348-x

Qianyu Song, Chao Li, Jinhu Fu, Qingtian Zeng, Nengfu Xie

Self-supervised heterogeneous graph neural networks have shown remarkable effectiveness in addressing the challenge of limited labeled data. However, current contrastive learning methods face limitations in leveraging neighborhood information for each node. Some approaches utilize the local information of the target node, ignoring useful signals from deeper neighborhoods. On the other hand, simply stacking convolutional layers to expand the neighborhood inevitably leads to over-smoothing. To address the problems, we propose HGNN-DB, a Self-supervised Heterogeneous Graph Neural Network Based on Deep and Broad Neighborhood Encoding to tackle the over-smoothing problem within heterogeneous graphs. Specifically, HGNN-DB aims to learn informative node representations by incorporating both deep and broad neighborhoods. We introduce a deep neighborhood encoder with a distance-weighted strategy to capture deep features of target nodes. Additionally, a single-layer graph convolutional network is employed for the broad neighborhood encoder to aggregate broad features of target nodes. Furthermore, we introduce a collaborative contrastive mechanism to learn the complementarity and potential invariance between the two views of neighborhood information. Experimental results on four real-world datasets and seven baselines demonstrate that our model significantly outperforms the current state-of-the-art techniques on multiple downstream tasks. The codes and datasets for this work are available at https://github.com/SSQiana/HGNN-DB.

{"title":"Self-supervised heterogeneous graph neural network based on deep and broad neighborhood encoding","authors":"Qianyu Song, Chao Li, Jinhu Fu, Qingtian Zeng, Nengfu Xie","doi":"10.1007/s10489-025-06348-x","DOIUrl":"10.1007/s10489-025-06348-x","url":null,"abstract":"<div><p>Self-supervised heterogeneous graph neural networks have shown remarkable effectiveness in addressing the challenge of limited labeled data. However, current contrastive learning methods face limitations in leveraging neighborhood information for each node. Some approaches utilize the local information of the target node, ignoring useful signals from deeper neighborhoods. On the other hand, simply stacking convolutional layers to expand the neighborhood inevitably leads to over-smoothing. To address the problems, we propose HGNN-DB, a Self-supervised <u>H</u>eterogeneous <u>G</u>raph <u>N</u>eural <u>N</u>etwork Based on <u>D</u>eep and <u>B</u>road Neighborhood Encoding to tackle the over-smoothing problem within heterogeneous graphs. Specifically, HGNN-DB aims to learn informative node representations by incorporating both deep and broad neighborhoods. We introduce a deep neighborhood encoder with a distance-weighted strategy to capture deep features of target nodes. Additionally, a single-layer graph convolutional network is employed for the broad neighborhood encoder to aggregate broad features of target nodes. Furthermore, we introduce a collaborative contrastive mechanism to learn the complementarity and potential invariance between the two views of neighborhood information. Experimental results on four real-world datasets and seven baselines demonstrate that our model significantly outperforms the current state-of-the-art techniques on multiple downstream tasks. The codes and datasets for this work are available at https://github.com/SSQiana/HGNN-DB.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 6","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465956","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}

引用次数: 0

A critical escape probability formulation for enhancing the transient stability of power systems with system parameter design

IF 4.8 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Automatica

Pub Date : 2025-02-22 DOI: 10.1016/j.automatica.2025.112217

Xian Wu , Kaihua Xi , Aijie Cheng , Chenghui Zhang , Hai Xiang Lin

For the enhancement of the transient stability of power systems, the key is to define a quantitative optimization formulation with system parameters as decision variables. In this paper, we model the disturbances by Gaussian noise and define a metric named Critical Escape Probability (CREP) based on the invariant probability measure of a linearized stochastic process. CREP characterizes the probability of the state escaping from a critical set. CREP involves all the system parameters and reflects the size of the basin of attraction of the nonlinear systems. An optimization framework that minimizes CREP with the system parameters as decision variables is presented. Simulations show that the mean of the first hitting time when the state hits the boundary of the critical set, that is often used to describe the stability of nonlinear systems, is dramatically increased by minimizing CREP. This indicates that the transient stability of the system is effectively enhanced. It is also shown that suppressing the state fluctuations only is insufficient for enhancing the transient stability. In addition, the famous Braess’ paradox which also exists in power systems is revisited. Surprisingly, it turned out that the paradoxes identified by the traditional metric may not exist according to CREP. This new metric opens a new avenue for the transient stability analysis of future power systems integrated with large amounts of renewable energy.

{"title":"A critical escape probability formulation for enhancing the transient stability of power systems with system parameter design","authors":"Xian Wu , Kaihua Xi , Aijie Cheng , Chenghui Zhang , Hai Xiang Lin","doi":"10.1016/j.automatica.2025.112217","DOIUrl":"10.1016/j.automatica.2025.112217","url":null,"abstract":"<div><div>For the enhancement of the transient stability of power systems, the key is to define a quantitative optimization formulation with system parameters as decision variables. In this paper, we model the disturbances by Gaussian noise and define a metric named <em>Critical Escape Probability</em> (CREP) based on the invariant probability measure of a linearized stochastic process. CREP characterizes the probability of the state escaping from a critical set. CREP involves all the system parameters and reflects the size of the basin of attraction of the nonlinear systems. An optimization framework that minimizes CREP with the system parameters as decision variables is presented. Simulations show that the mean of the first hitting time when the state hits the boundary of the critical set, that is often used to describe the stability of nonlinear systems, is dramatically increased by minimizing CREP. This indicates that the transient stability of the system is effectively enhanced. It is also shown that suppressing the state fluctuations only is insufficient for enhancing the transient stability. In addition, the famous Braess’ paradox which also exists in power systems is revisited. Surprisingly, it turned out that the paradoxes identified by the traditional metric may not exist according to CREP. This new metric opens a new avenue for the transient stability analysis of future power systems integrated with large amounts of renewable energy.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112217"},"PeriodicalIF":4.8,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143463382","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}

引用次数: 0

EEG-CNN-Souping: Interpretable emotion recognition from EEG signals using EEG-CNN-souping model and explainable AI

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-02-22 DOI: 10.1016/j.compeleceng.2025.110189

Eamin Chaudary, Sheeraz Ahmad Khan, Wajid Mumtaz

Emotion recognition is a key aspect of human–robot interaction (HRI), which requires social intelligence to perceive and react to human affective states. This paper introduces EEG-CNN-Souping, a novel approach that applies the “Model Soups” technique to a self-designed EEG-CNN model for classifying electroencephalogram (EEG) signals into emotions. EEG-CNN-Souping improves the model performance and efficiency by averaging the weights of multiple EEG-CNN models trained on different sizes of scalograms, which are acquired by applying continuous wavelet transform (CWT) and normalization to the EEG signals. The scalograms capture the time-varying patterns of the EEG signals effectively. The approach also uses data augmentation and gradient class activation map (Grad-Cam) visualization for robustness and interpretability respectively. The model is evaluated on a common dataset that is the SEED dataset and achieves a 99.31% accuracy, surpassing other state-of-the-art deep learning (DL) models in terms of accuracy, computational cost, and time efficiency. The prediction time for EEG-CNN-Souping is only 6 ms. The explainable artificial intelligence (XAI) method Grad-CAM is utilized for interpretation of predictions. EEG-CNN-Souping is computationally inexpensive and time-efficient.

{"title":"EEG-CNN-Souping: Interpretable emotion recognition from EEG signals using EEG-CNN-souping model and explainable AI","authors":"Eamin Chaudary, Sheeraz Ahmad Khan, Wajid Mumtaz","doi":"10.1016/j.compeleceng.2025.110189","DOIUrl":"10.1016/j.compeleceng.2025.110189","url":null,"abstract":"<div><div>Emotion recognition is a key aspect of human–robot interaction (HRI), which requires social intelligence to perceive and react to human affective states. This paper introduces EEG-CNN-Souping, a novel approach that applies the “Model Soups” technique to a self-designed EEG-CNN model for classifying electroencephalogram (EEG) signals into emotions. EEG-CNN-Souping improves the model performance and efficiency by averaging the weights of multiple EEG-CNN models trained on different sizes of scalograms, which are acquired by applying continuous wavelet transform (CWT) and normalization to the EEG signals. The scalograms capture the time-varying patterns of the EEG signals effectively. The approach also uses data augmentation and gradient class activation map (Grad-Cam) visualization for robustness and interpretability respectively. The model is evaluated on a common dataset that is the SEED dataset and achieves a 99.31% accuracy, surpassing other state-of-the-art deep learning (DL) models in terms of accuracy, computational cost, and time efficiency. The prediction time for EEG-CNN-Souping is only 6 ms. The explainable artificial intelligence (XAI) method Grad-CAM is utilized for interpretation of predictions. EEG-CNN-Souping is computationally inexpensive and time-efficient.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110189"},"PeriodicalIF":4.0,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Public Datasets for Cloud Computing: A Comprehensive Survey 云计算公共数据集：全面调查

IF 16.6 1区计算机科学 Q1 COMPUTER SCIENCE, THEORY & METHODS

ACM Computing Surveys

Pub Date : 2025-02-22 DOI: 10.1145/3719003

Guozhi Liu, Weiwei Lin, Haotong Zhang, Jianpeng Lin, Shaoliang Peng, Keqin Li

Publicly available datasets are vital to researchers because they permit the testing of new algorithms under a variety of conditions and ensure the verifiability and reproducibility of scientific experiments. In cloud computing research, there is a particular dependence on obtaining load traces and network traces from real cloud computing clusters, which are used for designing energy efficiency prediction, workload analysis, and anomaly detection solutions. To address the current lack of a comprehensive overview and thorough analysis of cloud computing datasets and to gain insight into their current status and future trends, in this paper, we provide a comprehensive survey of existing publicly cloud computing datasets. Firstly, we utilize a systematic mapping approach to analyze 968 scientific papers from 6 scientific databases, resulting in the retrieval of 42 datasets related to cloud computing. Secondly, we categorize these datasets based on 11 characteristics to assist researchers in quickly finding datasets suitable for their specific needs. Thirdly, we provide detailed descriptions of each dataset to assist researchers in gaining a clearer understanding of their characteristics. Fourthly, we select 12 mainstream datasets and conduct a comprehensive analysis and comparison of their characteristics. Finally, we discuss the weaknesses of existing datasets, identify challenges, provide recommendations for long-term dataset maintenance and updates, and outline directions for the future creation of new cloud computing datasets. Related resources are available at: https://github.com/ACAT-SCUT/Awesome-CloudComputing-Datasets.

{"title":"Public Datasets for Cloud Computing: A Comprehensive Survey","authors":"Guozhi Liu, Weiwei Lin, Haotong Zhang, Jianpeng Lin, Shaoliang Peng, Keqin Li","doi":"10.1145/3719003","DOIUrl":"https://doi.org/10.1145/3719003","url":null,"abstract":"Publicly available datasets are vital to researchers because they permit the testing of new algorithms under a variety of conditions and ensure the verifiability and reproducibility of scientific experiments. In cloud computing research, there is a particular dependence on obtaining load traces and network traces from real cloud computing clusters, which are used for designing energy efficiency prediction, workload analysis, and anomaly detection solutions. To address the current lack of a comprehensive overview and thorough analysis of cloud computing datasets and to gain insight into their current status and future trends, in this paper, we provide a comprehensive survey of existing publicly cloud computing datasets. Firstly, we utilize a systematic mapping approach to analyze 968 scientific papers from 6 scientific databases, resulting in the retrieval of 42 datasets related to cloud computing. Secondly, we categorize these datasets based on 11 characteristics to assist researchers in quickly finding datasets suitable for their specific needs. Thirdly, we provide detailed descriptions of each dataset to assist researchers in gaining a clearer understanding of their characteristics. Fourthly, we select 12 mainstream datasets and conduct a comprehensive analysis and comparison of their characteristics. Finally, we discuss the weaknesses of existing datasets, identify challenges, provide recommendations for long-term dataset maintenance and updates, and outline directions for the future creation of new cloud computing datasets. Related resources are available at: https://github.com/ACAT-SCUT/Awesome-CloudComputing-Datasets.","PeriodicalId":50926,"journal":{"name":"ACM Computing Surveys","volume":"26 1","pages":""},"PeriodicalIF":16.6,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143470934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Learning discriminative features for multi-hop knowledge graph reasoning

IF 3.4 2区计算机科学 Q2 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Applied Intelligence

Pub Date : 2025-02-22 DOI: 10.1007/s10489-025-06327-2

Hao Liu, Dong Li, Bing Zeng, Yang Xu

Reinforcement learning (RL)-based multi-hop knowledge graph reasoning has achieved remarkable success in real-world applications and can effectively handle knowledge graph completion tasks. This approach involves a policy-based agent navigating the graph environment to extend reasoning paths and identify the target entity. However, most existing multi-hop reasoning models are typically constrained to stepwise inference, which inherently disrupts the global information of multi-hop paths. To overcome this limitation, we introduce discriminative features between valid and invalid paths as global information. Here, we propose a multi-hop path encoder specifically designed to extract these discriminative features. Firstly, a multi-hop path encoding module is employed to derive each path’s hidden features, using cross-attention mechanisms to strengthen the interaction between triple context and path features. Secondly, a discriminative feature extraction module is used to capture the differences between valid and invalid paths. Thirdly, an attention-enhanced gated fusion mechanism is implemented to integrate these discriminative features into the multi-hop inference decoder. We further evaluate our method on five standard datasets. Our method outperforms the baseline models, demonstrating the effectiveness of discriminative features in improving prediction performance, learning speed, and path interpretability.

{"title":"Learning discriminative features for multi-hop knowledge graph reasoning","authors":"Hao Liu, Dong Li, Bing Zeng, Yang Xu","doi":"10.1007/s10489-025-06327-2","DOIUrl":"10.1007/s10489-025-06327-2","url":null,"abstract":"<div><p>Reinforcement learning (RL)-based multi-hop knowledge graph reasoning has achieved remarkable success in real-world applications and can effectively handle knowledge graph completion tasks. This approach involves a policy-based agent navigating the graph environment to extend reasoning paths and identify the target entity. However, most existing multi-hop reasoning models are typically constrained to stepwise inference, which inherently disrupts the global information of multi-hop paths. To overcome this limitation, we introduce discriminative features between valid and invalid paths as global information. Here, we propose a multi-hop path encoder specifically designed to extract these discriminative features. Firstly, a multi-hop path encoding module is employed to derive each path’s hidden features, using cross-attention mechanisms to strengthen the interaction between triple context and path features. Secondly, a discriminative feature extraction module is used to capture the differences between valid and invalid paths. Thirdly, an attention-enhanced gated fusion mechanism is implemented to integrate these discriminative features into the multi-hop inference decoder. We further evaluate our method on five standard datasets. Our method outperforms the baseline models, demonstrating the effectiveness of discriminative features in improving prediction performance, learning speed, and path interpretability.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 6","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465957","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}

引用次数: 0

首页上一页

下一页尾页

类型

计算机科学最新文献

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

全部 J. Field Rob. J. Bionic Eng. ACTA INFORM Adv. Rob. AI MAG Ann. Math. Artif. Intell. Appl. Bionics Biomech. APPL INTELL APPL COMPUT ELECTROM APPL ARTIF INTELL Artif. Intell. ARTIF INTELL REV CHEMOMETR INTELL LAB China Commun. CMC-Comput. Mater. Continua Complex Intell. Syst. Comput. Sci. Eng. Commun. ACM COMPUTER Comput. Graphics Forum COMPUTING EMPIR SOFTW ENG Enterp. Inf. Syst. EPJ Data Sci. ETRI J EURASIP J WIREL COMM Evolving Systems FORM METHOD SYST DES Front. Neurorob. FRONT COMPUT SCI-CHI IEEE Trans. Commun. IEEE Trans. Comput. Social Syst. IEEE Trans. Dependable Secure Comput. IEEE Trans. Green Commun. Networking IEEE Trans. Cognit. Commun. Networking IEEE Access IEEE Trans. Comput. IEEE Antennas Propag. Mag. IEEE Micro IEEE Trans. Antennas Propag. IEEE Trans. Control Syst. Technol. IEEE Trans. Big Data IEEE Trans. Cybern. IEEE Internet Comput. IEEE Trans. Affective Comput. IEEE Trans. Emerging Top. Comput. Intell. IEEE SECUR PRIV IEEE Trans. Emerging Top. Comput. IEEE Trans. Aerosp. Electron. Syst. IEEE Trans. Broadcast. IEEE Intell. Syst. IEEE Commun. Lett. IEEE Trans. Autom. Control IEEE Trans. Cloud Comput. IEEE Trans. Evol. Comput. IEEE Trans. Consum. Electron. IEEE Trans. Fuzzy Syst. IEEE Trans. Haptic IEEE Trans. Image Process. IEEE Multimedia IEEE Rob. Autom. Lett. IEEE J. Sel. Areas Commun. IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. IETE Tech. Rev. IEEE Trans. Serv. Comput. IEEE Trans. Parallel Distrib. Syst. IEEE Trans. Sustainable Comput. IEEE Trans. Multimedia IEEE Trans. Ind. Inf. IEEE Trans. Neural Networks Learn. Syst. IEEE Trans. Software Eng. IEEE-ACM T AUDIO SPE IEEE Wireless Commun. IEEE Wireless Commun. Lett. IET MICROW ANTENNA P IEEE Trans. Visual Comput. Graphics IEEE Trans. Ind. Electron. IET Optoelectron IEEE Trans. Veh. Technol. IEEE Trans. Netw. Serv. Manage. IEEE Trans. Pattern Anal. Mach. Intell. IEEE Trans. Wireless Commun. IEEE ACM T NETWORK IEEE Trans. Inf. Forensics Secur. IEEE Trans. Inf. Theory IEEE Trans. Knowl. Data Eng. INFORM SYST FRONT INFORMS J COMPUT INFOR Int. J. Comput. Vision Int. J. Approximate Reasoning Int. J. Control Int. J. Commun. Syst. Int. J. Imaging Syst. Technol. Int. J. Fuzzy Syst. Int. J. Intell. Syst. Int. J. Network Manage. Int. J. Parallel Program. Int. J. Social Rob. Int. J. Software Tools Technol. Trans.

﹀