Article
Computer Science, Artificial Intelligence
Sara Hosseinzadeh Kassani, Farhood Rismanchian, Peyman Hosseinzadeh Kassani
Summary: This study combines k-nearest neighbor rule and relevance vector machines, proposing a k relevance vector model (k-RV) to optimize performance by selecting important features and considering relevancy in the feature space. Introducing a new parameter for improving classification accuracy, the k-RV model shows competitive performance in experiments compared to state-of-the-art methods.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Mechanical
R. Fuentes, R. Nayek, P. Gardner, N. Dervilis, T. Rogers, K. Worden, E. J. Cross
Summary: This paper presents a new Bayesian approach to equation discovery in nonlinear structural dynamics, combining structure detection and parameter estimation. By using a sparsity-inducing prior and an over-complete dictionary, the method successfully identifies and validates equations for nonlinear dynamic systems. Unlike other sparse learners, this approach utilizes hierarchical Bayesian priors and hyperpriors to achieve accurate results.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Mir Jafar Sadegh Safari, Shervin Rahimzadeh Arashloo
Summary: The study utilizes sparse kernel regression (SKR) technique to design a self-cleaning sediment transport model, with the comparison against support vector regression (SVR) showing SKR's superior efficacy in generating accurate results across a wide range of channel characteristics.
ADVANCED ENGINEERING INFORMATICS
(2021)
Article
Robotics
Thai Duong, Michael Yip, Nikolay Atanasov
Summary: This article discusses the use of online occupancy mapping and real-time collision checking for autonomous robot navigation in unknown environments. A new approach is proposed, which utilizes machine learning classifiers to efficiently determine the boundary between occupied and free space. The effectiveness of the proposed methods is evaluated through tasks involving autonomous navigation and mapping in unknown environments.
IEEE TRANSACTIONS ON ROBOTICS
(2022)
Article
Computer Science, Artificial Intelligence
Jiahua Luo, Yanfen Gan, Chi-Man Vong, Chi-Man Wong, Chuangquan Chen
Summary: Sparse Bayesian learning (SBL) excels in accuracy, sparsity, and probabilistic prediction for classification. However, its scalability to large problems is hindered by the inversion of a potentially enormous covariance matrix. This paper introduces an approximate SBL algorithm called ARP-SBL, which addresses this scalability issue by approximating regularization priors without the need for inverting the covariance matrix.
Article
Engineering, Civil
Yuan-Hao Wei, You-Wu Wang, Yi-Qin Ni
Summary: The study aims to develop a wheel defect detection approach based on RVM that can detect defects online using trackside monitoring data under different running-speed conditions. By extracting CFA from dynamic strain responses and formulating multiple probabilistic regression models (MPRMs) using multi-kernel RVM, the proposed approach demonstrates better local and global representation ability and generalization performance for reliable defect detection. The method is validated using real-world monitoring data acquired by an FBG-based trackside monitoring system, showing its effectiveness under different running-speed conditions.
SMART STRUCTURES AND SYSTEMS
(2022)
Article
Management
Rafael Blanquero, Emilio Carrizosa, Cristina Molero-Rio, Dolores Romero Morales
Summary: This paper proposes an optimal regression tree model based on a continuous optimization problem, which aims to strike a balance between prediction accuracy and sparsity. The model can fulfill important properties for regression tasks and provide local explanations due to the smoothness of predictions. The computational experience demonstrates the superiority of this approach in terms of prediction accuracy compared to standard benchmark regression methods, and the scalability with respect to sample size is also illustrated.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Computer Science, Interdisciplinary Applications
Tyler Buffington, James G. Scott, Ofodike A. Ezekoye
Summary: This study explores different spatial and sociodemographic models to predict residential fire counts in census tracts for 118 U.S. fire departments across 25 states. The research highlights the impact of socioeconomic factors on fire risk and proposes a Bayesian hierarchical Poisson regression model that improves prediction accuracy.
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Zhiwang Zhang, Hongliang Sun, Shuqing Li, Jing He, Jie Cao, Guanghai Cui, Gang Wang
Summary: This paper proposes a novel two-stage sparse multi-kernel optimization classifier (TSMOC) method to address the issue of inconsistent classification caused by redundancy and unrelated attributes. By introducing single or multi-kernel functions into classifier models, non-linearly separable problems are solved, but predictive interpretability is reduced. Experimental results show that TSMOC outperforms seven other classifiers on thirteen real-world datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Luca Martino, Jesse Read
Summary: This article introduces and discusses Gaussian Processes and Relevance Vector Machines, which straddle probabilistic Bayesian schemes and kernel methods. The focus is on developing a common framework, drawing connections among them via dual formulations, and discussing their application in major tasks such as regression, smoothing, interpolation, and filtering. It provides understanding of the mathematical concepts behind these models, summarizing and discussing different interpretations, and highlighting their relationships to other methods.
INFORMATION FUSION
(2021)
Article
Engineering, Electrical & Electronic
Long Chen, Jun Zhao, Wei Wang, Qingshan Xu
Summary: This study introduces an RVM prediction model with input noise to handle input uncertainty, utilizes a Gaussian approximation for input uncertainty, and employs the Markov chain Monte Carlo algorithm to approximate the posterior distribution over model weights, resulting in improved prediction performance.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Environmental Sciences
Jiayi Zhao, Xiaoyue Shi, Zhiqin Wang, Sijie Xiong, Yongfeng Lin, Xiaoran Wei, Yanwei Li, Xiaowen Tang
Summary: Per- and polyfluoroalkyl substances (PFASs) are persistent organic pollutants that have been detected in various environmental media and human serum. This study evaluated the binding affinity of PFASs to liver fatty acid binding protein (L-FABP) and used the data to develop a machine learning model for predicting potentially hazardous PFASs. The study also found that flexibility is an important molecular property in PFASs-induced hepatotoxicity.
ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY
(2023)
Article
Computer Science, Information Systems
Wenyang Wang, Dongchu Sun, Peng Shao, Haibo Kuang, Cong Sui
Summary: This paper introduces a new RVM classification model PRVM, which replaces the logistic link function with the probit link function, simplifies Bayesian computation by introducing a latent variable, and provides a closed-form solution. Compared to the original model, the new model is a Fully Bayesian approach with improved computational efficiency. Different prior structures have an impact on model performance.
Article
Engineering, Civil
C. Narendra Babu, Pallaviram Sure, Chandra Mohan Bhuma
Summary: Real-time road network traffic state estimation is crucial for enhancing Intelligent Transportation Systems (ITS). This study proposes sparse representation methods using Sparse Bayesian Learning (SBL) and Block SBL (BSBL) to address data vacancies and improve accuracy. By leveraging historical spatio-temporal correlations and kalman filtering, the proposed approach achieves less than 6% Normalized Mean Absolute Error (NMAE) in experiments with PeMS traffic data, demonstrating its effectiveness for online traffic state estimation.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Thermodynamics
Tiago de Oliveira Nogueira, Gilderlanio Barbosa Alves Palacio, Fabricio Damasceno Braga, Pedro Paulo Nunes Maia, Elineudo Pinho de Moura, Carla Freitas de Andrade, Paulo Alexandre Costa Rocha
Summary: This study combines DFA with SVM and RBFK methods for imbalance level classification in wind turbines, with RBFK showing excellent performance in different rotation speeds.