检索结果-内蒙古大学图书馆

CardiOT: Towards Interpretable Drug Cardiotoxicity prediction Using Optimal Transport and Kolmogorov--Arnold Networks

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2025年第3期29卷 1759-1770页

作者： Zhang, Xinyu Wang, Hao Du, Zhenya Zhuo, Linlin Fu, Xiangzheng Cao, Dongsheng Xie, Boqia Li, Keqin Wenzhou Univ Technol Sch Data Sci & Artificial Intelligence Wenzhou 325027 Peoples R China Chinese Acad Agr Sci Inst Crop Sci Beijing 100081 Peoples R China Guangzhou Xinhua Univ Guangzhou 510520 Peoples R China Hunan Univ Coll Informat Sci & Engn Changsha 410082 Peoples R China Cent South Univ Xiangya Sch Pharmaceut Sci Changsha 410003 Peoples R China Capital Med Univ Beijing Chaoyang Hosp Cardiovasc Imaging Ctr Dept Cardiol Beijing 100020 Peoples R China State Univ New York Dept Comp Sci New York NY 12561 USA

Investigating the inhibitory effects of compounds on cardiac ion channels is essential for assessing cardiac drug safety. Consequently, researchers have developed computational models to evaluate combined cardiotoxicity (CCT) on cardiac ion channels. However, limitations in experimental data often cause issues like uneven data distribution and scarcity. Additionally, existing models primarily emphasize atomic information flow within graph neural networks (GNNs) while overlooking chemical bonds, leading to inadequate recognition of key structures. Therefore, this study integrates optimal transport (OT), structure remapping (SR), and Kolmogorov-Arnold networks (KANs) into a GNN-based CCT prediction model, CardiOT. First, the proposed CardiOT model employs OT pooling to optimize sample-feature joint distribution using expectation maximization, identifying "important" sample-feature pairs. Additionally, SR technology is used to emphasize the role of chemical bond information in message propagation. KAN technology is integrated to greatly enhance model interpretability. In summary, the model mitigates challenges related to uneven data distribution and scarcity. Multiple experiments on public datasets confirm the model's robust performance. We anticipate that this model will provide deeper insights into compound inhibition mechanisms on cardiac ion channels and reduce toxicity risks.

关键词： Drugs Compounds Predictive models Ions Chemicals Safety Data models Accuracy Toxicology prediction algorithms Cardiac drug safety compound cardiotoxicity (CCT) inhibitory effects of drugs Kolmogorov--Arnold networks (KANs) optimal transport (OT) uneven distribution and data scarcity

来源：评论

学校读者我要写书评

暂无评论

Stock Market prediction Using a Hybrid Approach With Boruta and Liquid Neural Network

引用

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS 2025年

作者： Reddy, G. Joshita Sahani, Rahul Kumar Raja, S. P. Vellore Inst Technol Dept Comp Sci & Engn Vellore 632014 India

Globally, one of the most fundamental applications of time-series data is stock market forecasting, where each and every second is crucial and analysis is unpredictable, posing a significant challenge towards predicting its trend. Various research is underway for the analysis of the stock market using various time series algorithms such as long-short-term memory (LSTM), recurrent neural network (RNN). Here, in this article, we have proposed a novel approach of multiple layers including liquid time constant, the linear first order dynamics with nonlinear interlinked gates which improves performance of time series;after which we included liquid neural network and forecasted the results. Then, we enhanced the algorithm using Boruta feature selection, where we have selected only the required columns and ranked them as per requirement. Furthermore, for more accurate results, we have selected three different sets of historical data for five large cap sectors (Alphabet, Apple, Blackrock, JP Morgan, and IBM) after which we observed the R-square values along with all the performance evaluation and error evaluation results for all the individual dataset in two phases, i.e., before and after using the Boruta feature selection method for 5, 10, and 15 years data, we observed the best results for LSTM (0.9443) and RNN (0.9752). Liquid neural networks, being the most efficient and accurate, gave the best R-square value of 0.9835.

关键词： Stock markets Companies Long short term memory prediction algorithms Liquids Market research Predictive models Time series analysis Heuristic algorithms Data models Boruta algorithm liquid neural networks (LNNs) liquid time constant (LTC) stock market prediction

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Interval Auto-Scaling Optimization Method Based on Informer Time Series prediction

引用

IEEE ACCESS 2025年 13卷 14572-14583页

作者： Ding, Yu Li, Chenhao Cai, Zhengong Wang, Xinghao Yang, Bowei Zhejiang Univ Coll Comp Sci & Technol Hangzhou 310027 Zhejiang Peoples R China Zhejiang Univ Sch Software Technol Ningbo 315048 Zhejiang Peoples R China Zhejiang Univ Sch Aeronaut & Astronaut Hangzhou 310012 Zhejiang Peoples R China

With the rapid development and application of container cloud computing-related technologies, more and more applications are being deployed to container cloud clusters. As an essential feature of container cloud platforms and cloud-native architecture, auto-scaling aims to automatically and quickly adjust the allocation of cloud resources according to the resource requirements of applications. Currently, widely used responsive auto-scaling methods, such as Kubernetes HPA, exhibit certain lags due to the startup time costs of containers and Pods. This lag makes it difficult to guarantee the service quality of applications when there is a sudden increase in online application load. This paper proposes a dynamic interval auto-scaling optimization method based on Informer time series prediction. By predicting online application load and dynamically determining the auto-scaling interval, sufficient resources are allocated to the application in advance. In the experiments conducted on the official World Cup forum load and Alibaba cluster CPU load, the Informer time series prediction algorithm demonstrated better long-sequence time series prediction capabilities compared to algorithms such as LSTM and RNN. In elastic scaling experiments, compared to Kubernetes HPA, the method proposed in this paper reduces the average application response time from 0.821 seconds to 0.692 seconds, and the SLA violation rate decreases from 18.277% to 9.157%. This indicates a significant improvement in the service quality metrics of online applications. Furthermore, the proposed method effectively maintains a balance between high CPU resource utilization and low application response time and SLA violation rate, which is something RNN-based elastic scaling method cannot achieve, as it can only reduce application response time and SLA violation rate by sacrificing CPU resource utilization.

关键词： Containers Cloud computing Time series analysis Long short term memory Load modeling Resource management prediction algorithms Measurement Autoregressive processes Predictive models Auto-scaling container cloud dynamic interval informer time series prediction

来源：评论

学校读者我要写书评

暂无评论

Enhanced Workload prediction in Data Centers Using Two-Stage Decomposition and Hybrid Parallel Deep Learning

引用

IEEE ACCESS 2025年 13卷 64115-64132页

作者： Alqahtani, Dalal Imani, Hamidreza El-Ghazawi, Tarek George Washington Univ Dept Elect & Comp Engn Washington DC 20052 USA Najran Univ Coll Comp Sci & Informat Syst Dept Comp Sci Najran 61441 Saudi Arabia

Workload prediction is one of the most basic requirements in developing cost and energy-efficient Cloud Data Centers (CDCs). Most traditional approaches have suffered from noise and failed to capture the complex dynamic patterns in workload data, reducing their accuracy. To improve this, we introduce CVCBM which blends signal processing techniques Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Variational Mode Decomposition (VMD) with advanced deep learning models like Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) networks. CVCBM utilizes a hierarchical two-stage decomposition process, beginning with CEEMDAN. We use the capability of CEEMDAN for denoising by additive white noise to clean the noisy workload data and further decompose it into several Intrinsic Mode Functions (IMFs), ranging from high to low frequencies. Then, we propose a partitional clustering approach based on Sample Entropy (SE) to select components of similar complexity to increase the effectiveness of the second-stage denoising. After that, VMD is a method based on the center frequency applied to decompose further the high-frequency components, which may contain fluctuations that obscure underlying trends, thus enhancing the model's overall accuracy. A novel hybrid model is utilized to forecast future workloads, incorporating two sets of three distinct parallel Conv1D layers with varying kernel sizes. These layers extract patterns from the input data, capturing short-term, medium-term, and long-term workload information, allowing the model to learn variations at different scales. Following this, Bi-LSTM layers capture the temporal dependencies within the patterns identified by the Conv1D layers. Extensive experiments on various real-world datasets from Google and Alibaba show that CVCBM significantly outperforms the other methods, which makes it a solution for future workload prediction in cloud data centers.

关键词： Time series analysis Predictive models Data centers Cloud computing Accuracy Feature extraction Deep learning Resource management Data models prediction algorithms Cloud data center workload prediction deep learning hybrid prediction data decomposition

来源：评论

学校读者我要写书评

暂无评论

Fast and Accurate Multi-Agent Trajectory prediction for Crowded Unknown Scenes

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 7606-7621页

作者： Tao, Xiuye Li, Huiping Liang, Bin Shi, Yang Xu, Demin Northwestern Polytech Univ Sch Marine Sci & Technol Xian 710072 Peoples R China Tsinghua Univ Dept Automat Beijing 100084 Peoples R China Univ Victoria Dept Mech Engn Victoria BC V8P 5C2 Canada

This paper studies the problem of multi-agent trajectory prediction in crowded unknown environments. A novel energy function optimization-based framework is proposed to generate prediction trajectories. Firstly, a new energy function is designed for easier optimization. Secondly, an online optimization pipeline for calculating parameters and agents' velocities is developed. In this pipeline, we first design an efficient group division method based on Frechet distance to classify agents online. Then the strategy on decoupling the optimization of velocities and critical parameters in the energy function is developed, where the slap swarm algorithm and gradient descent algorithms are integrated to solve the optimization problems more efficiently. Thirdly, we propose a similarity-based resample evaluation algorithm to predict agents' optimal goals, defined as the target-moving headings of agents, which effectively extracts hidden information in observed states and avoids learning agents' destinations via the training dataset in advance. Experiments and comparison studies verify the advantages of the proposed method in terms of prediction accuracy and speed.

关键词： Trajectory Predictive models Accuracy Robots prediction algorithms Real-time systems Navigation Long short term memory Service robots Learning systems Trajectory prediction crowded environments energy function Frechet distance similarity-based resample evaluation algorithm

来源：评论

学校读者我要写书评

暂无评论

A Multisource Signal-Conditioned Fiber Optic Sensing System for Microwave Temperature Acquisition and prediction

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Meng, Chenyang Hao, Kuangrong Cheng, Yan Xu, Jinhai Zhu, Xiuli Donghua Univ Coll Informat Sci & Technol Shanghai 201620 Peoples R China Univ Shanghai Sci & Technol Coll Opt Elect & Comp Engn Shanghai 200093 Peoples R China

Due to the existence of electromagnetic standing waves, uniform heating and accurate temperature data collection during microwave cooking have always been a challenge. This article presents a multisource temperature data acquisition system for microwave heating environments and introduces a brain-inspired prediction algorithm based on data structure to enhance temperature control logic. First, a temperature monitoring system for microwave heating environments, termed fiber optic temperature sensing system (FOTS), is developed. This system consists of an eight-probe fiber optic temperature sensor (FOT) array and an infrared temperature sensor. Then, a brain-inspired audiovisual (BIAV) predictive algorithm is proposed to simulate the brain's integration mechanism. The core of this algorithm extracts key time series data, calculates trend features to output the related key moment labels, and puts forward a vision-enhanced LSTM for prediction. Finally, experiments demonstrate the accuracy of BIAV in predicting food temperatures using sensor data, and existing control algorithms are optimized and validated. The proposed FOTS and BIAV algorithm in microwave fields provide new ideas and methods for multisensor data processing. In addition, existing control algorithms are optimized using predicted accurate temperature values and 2-D temperature fields, making microwave heating more uniform and safer, which is of great significance for industrial control.

关键词： Temperature sensors Temperature measurement Temperature distribution Electromagnetic heating Optical fibers Microwave measurement prediction algorithms Microwave ovens Heating systems Accuracy Automatic temperature control algorithm brain inspiration fiber optic temperature sensing system multisensor data temperature time series data prediction

来源：评论

学校读者我要写书评

暂无评论

Adaptive prediction Ensemble: Improving Out-of-Distribution Generalization of Motion Forecasting

引用

IEEE ROBOTICS AND AUTOMATION LETTERS 2025年第2期10卷 1553-1560页

作者： Li, Jinning Li, Jiachen Bae, Sangjae Isele, David Univ Calif Berkeley Berkeley CA 94720 USA Univ Calif Riverside Riverside CA 92521 USA Honda Res Inst San Jose CA 95134 USA

Deep learning-based trajectory prediction models for autonomous driving often struggle with generalization to out-of-distribution (OOD) scenarios, sometimes performing worse than simple rule-based models. To address this limitation, we propose a novel framework, Adaptive prediction Ensemble (APE), which integrates deep learning and rule-based prediction experts. A learned routing function, trained concurrently with the deep learning model, dynamically selects the most reliable prediction based on the input scenario. Our experiments on large-scale datasets, including Waymo Open Motion Dataset (WOMD) and Argoverse, demonstrate improvement in zero-shot generalization across datasets. We show that our method outperforms individual prediction models and other variants, particularly in long-horizon prediction and scenarios with a high proportion of OOD data. This work highlights the potential of hybrid approaches for robust and generalizable motion prediction in autonomous driving.

关键词： Predictive models Training Routing prediction algorithms Trajectory Adaptation models Deep learning Reliability Autonomous vehicles Decoding Intelligent transportation systems deep learning methods

来源：评论

学校读者我要写书评

暂无评论

prediction of High-Pressure Physical Properties of Crude Oil Using Explainable Machine Learning Models

引用

IEEE ACCESS 2025年 13卷 4912-4931页

作者： He, Yawen Chen, Jianwen Wu, Zhiyu Dun, Yaxin Du, Guichao Feng, Mengsen Lu, Yifan Dang, Wei Northwest Univ Dept Geol Xian 710069 Peoples R China Northwest Univ State Key Lab Continental Dynam Xian 710069 Peoples R China PetroChina Changqing Oilfield Co Oil Prod Plant 12 Heshui 745400 Peoples R China Xian Shiyou Univ Sch Earth Sci & Engn Xian 710069 Peoples R China

High-pressure physical property parameters of formation crude oil are crucial for oilfield exploration, development, and production. Various prediction models have been developed using PVT experimental methods, empirical formulas, regression analysis, and machine learning techniques. However, these models often lack sufficient transparency and clarity in their computational processes, limiting their effectiveness in oil and gas exploration and development. This paper employs the Extreme Gradient Boosting (XGBoost) machine learning algorithm, utilizing experimental data such as formation pressure, formation temperature, surface crude oil density, natural gas relative density, gas-oil ratio, saturation pressure, volume factor, formation crude oil density, and formation crude oil viscosity as the dataset. The machine learning model is trained with these parameters to develop the XGBoost prediction model. Optimal parameters, including gas-oil ratio, saturation pressure, volume factor, formation crude oil density, and formation crude oil viscosity, are selected, and the Particle Swarm Optimization (PSO) algorithm is applied to further optimize the model, resulting in the PSO-XGBoost prediction model. To enhance the understanding of the model's operation and the influencing factors, Shapley Additive Explanations (SHAP) and Explainable Boosting Machine (EBM) techniques are employed for both local and global interpretations of the machine learning model. This study identifies key factors and their weights in predicting high-pressure physical property parameters of formation crude oil, providing valuable insights for predicting shale oil reservoir parameters and enhancing exploration and development efficiency.

关键词： Oils Predictive models Reservoirs Viscosity Boosting Artificial neural networks Data models Computational modeling Accuracy prediction algorithms Extreme gradient boosting particle swarm optimization model interpretation machine learning PVT property

来源：评论

学校读者我要写书评

暂无评论

Offender Characterization and prediction: A Case Study of the Kingdom of Bahrain

引用

IEEE ACCESS 2025年 13卷 29406-29431页

作者： Alasfoor, Ebrahim Jaffar Alshaikh, Omar Inuwa-Dutse, Isa Khan, Saad Parkinson, Simon Minist Interior Royal Acad Police Manama 973 Bahrain Univ Huddersfield Dept Comp Sci Huddersfield HD1 3DH England

Crime analysis using data mining techniques can uncover valuable information to assist law enforcement investigations. This study analyses a crime data set from the Kingdom of Bahrain to demonstrate an integrated framework using exploratory analysis and machine learning for predictive modelling. The study data set, comprising 720 cases, was first explored to reveal patterns, including prevalent theft and assault crimes concentrated in four areas. Data sensitivity measures were implemented by anonymising specific geographic areas, represented by characters A to T. Supervised learning algorithms such as Additive Regression, Random Forest, Na & iuml;ve Bayes, and Decision Tree are then applied to predict unknown offender characteristics such as age, sex, and occupation. The algorithms achieve strong predictive performance. Additive regression achieved a high correlation coefficient of 0.889 for the prediction of the age of the offender, and Random Forest achieved 83% precision in predicting sex through 10-fold cross-validation. However, challenges arise with unbalanced nationality data. A survey of 22 police officers was conducted, discussing the study results and the feasibility of the predictive modelling approach to enhance investigative efficiency. The responses received were mostly positive, reflecting a strong endorsement of the potential of the proposed method to enhance law enforcement practices. This research highlights a promising methodology that combines exploratory and machine learning techniques to extract actionable insights from crime data. With appropriate caution, it can improve investigative efficiency by making data-driven predictions of the attributes of the offender. Further work should expand data sets and explore advanced algorithms while addressing ethical concerns regarding predictive modelling in criminal justice.

关键词： Data mining Law enforcement Machine learning algorithms Accuracy prediction algorithms Predictive models Prevention and mitigation Data models Random forests Analytical models Crime analysis criminal investigation data mining machine learning predictive modelling EDA

来源：评论

学校读者我要写书评

暂无评论

Vehicular Speed prediction Method for Highway Scenarios Based on Spatiotemporal Graph Convolutional Networks and Potential Field Theory

引用

IEEE INTERNET OF THINGS JOURNAL 2025年第3期12卷 3330-3349页

作者： Li, Linheng An, Bocheng Zhang, Dapeng Gan, Rui Zhou, Zhi Qu, Xu Ran, Bin Southeast Univ Sch Transportat Nanjing 211189 Peoples R China Southeast Univ Inst Internet Mobil Nanjing 211189 Peoples R China Southeast Univ Univ Wisconsin Madison Nanjing 211189 Peoples R China Southeast Univ Jiangsu Prov Collaborat Innovat Ctr Modern Urban T Nanjing 211189 Peoples R China Southwestern Univ Finance & Econ Sch Management Sci & Engn Chengdu 611100 Peoples R China

Traffic flow analysis largely depends on accurate predictions of microscopic speed. Due to the complexity and stochastic of real-world driving environments, traditional model-driven methods face significant challenges. In recent years, data-driven methods that combine advanced intelligent algorithms have emerged to address the issue of vehicle speed prediction. However, most existing studies primarily focus on the static spatiotemporal relationships between vehicles, with only a few exploring dynamic spatiotemporal correlations. In this study, a new graph structure, named the vehicle generalized-dynamic graph, is constructed to characterize the vehicle trajectory, emphasizing the spatial and temporal evolution of vehicles. In the spatial dimension, we employ the potential field theory to compute the field strength of vehicles in different motion states, enhancing the description of interdependencies among vehicles. In the temporal dimension, the movements of each vehicle at different time points are interrelated. Furthermore, we propose a novel neural network architecture, named the dynamic edge graph convolutional network (DEGCN), to address the vehicle speed prediction problem. The DEGCN model is evaluated through experiments on three real-world vehicle trajectory datasets. The experimental results demonstrate that the proposed model outperforms other baseline models in terms of prediction performance. Additionally, various ablation studies are conducted to evaluate the effectiveness of different components, the potential field theory, and the vehicle generalized-dynamic graph structure.

关键词： Predictive models Data models Microscopy Vehicle dynamics Dynamics Trajectory Accuracy prediction algorithms Graph convolutional networks Force Graph convolutional network (GCN) potential field theory traffic flow vehicle generalized-dynamic graph vehicle speed prediction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：