Article
Computer Science, Artificial Intelligence
M. Tanveer, A. Tiwari, R. Choudhary, M. A. Ganaie
Summary: This study proposes a novel large scale pinball twin support vector machine (LPTWSVM) to address the limitations of the twin support vector machines (TWSVMs), using a unique pinball loss function and improving model performance by eliminating matrix inversion calculation and minimizing structural risk.
Article
Computer Science, Artificial Intelligence
Chun-Na Li, Yuan-Hai Shao, Huajun Wang, Yu-Ting Zhao, Naihua Xiu, Nai-Yang Deng
Summary: This paper investigates the general forms and characteristics of nonparallel support vector machines (NSVMs) and categorizes them into two types. It reveals the advantages and defects of different types and points out the inconsistency problems. Based on this observation, a novel max-min distance-based NSVM is proposed with desired consistency. The proposed NSVM has the consistency of training and test and the consistency of metric, and it assigns each sample an ascertained loss.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaoming Wang, Shitong Wang, Zengxi Huang, Yajun Du
Summary: This paper introduces a novel method called sparse support vector machine guided by radius-margin bound (RMB-SSVM) to efficiently condense the basis vectors in support vector machines. By selecting basis vectors and learning corresponding coefficients with a criterion related to SVM's generalization ability, the RMB-SSVM model can yield better performance.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Sambhav Jain, Reshma Rastogi
Summary: This paper proposes Parametric non-parallel support vector machines for binary pattern classification. The model brings noise resilience and sparsity by intelligently redesigning the Support vector machine optimization. The experimental results validate its scalability for large scale problems.
Article
Computer Science, Information Systems
Sebastian Maldonado, Julio Lopez, Carla Vairetti
Summary: The predictive performance of classification methods relies heavily on the nature of the environment and dataset shift issue. A novel Fuzzy Support Vector Machine strategy is proposed in this paper to improve performance by redefining the loss function and applying aggregation operators to deal with dataset shift. Our methods outperform traditional classifiers in terms of out-of-time prediction using simulated and real-world dataset for credit scoring.
INFORMATION SCIENCES
(2021)
Article
Chemistry, Analytical
Simone Carone, Giovanni Pappalettera, Caterina Casavola, Simone De Carolis, Leonardo Soria
Summary: Machine learning techniques combined with machine condition monitoring can diagnose faults more accurately than other condition-based monitoring approaches. However, statistical or model-based methods are often not applicable in highly customized industrial environments. Monitoring the health of bolted joints is crucial for maintaining structural integrity. Yet, there is little research on detecting bolt loosening in rotating joints.
Article
Computer Science, Artificial Intelligence
Wangyong Lv, Tingting Li, Huali Ren, Shijing Zeng, Jiao Zhou
Summary: The IDH-MSVM algorithm adjusts the distance between hyperplanes and classical margins to handle multiclassification problems more flexibly. Experimental results on UCI standard data sets show that this method achieves better classification accuracy for multiclass data compared to other algorithms.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Mathematics
Roberto Barcenas, Maria Gonzalez-Lima, Joaquin Ortega, Adolfo Quiroz
Summary: The effectiveness of subsampling methods in reducing the required instances in the training stage of using support vector machines (SVMs) for classification in big data scenarios is explored in this paper with theoretical results. The main theorem states that, under certain conditions, a feasible solution can be found for the SVM problem using a randomly chosen subsample, which can be as close as desired to the classifier trained with the complete dataset in terms of classification error. Additionally, a new subsampling method called importance sampling and bagging is proposed, which provides a faster solution to the SVM problem without significant loss in accuracy compared to existing techniques.
Article
Computer Science, Theory & Methods
Sandamal Weerasinghe, Tansu Alpcan, Sarah M. Erfani, Christopher Leckie
Summary: The research shows that with a new defense algorithm and metric method, SVMs can improve resistance against targeted attacks and significantly reduce classification error rates.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2021)
Article
Computer Science, Artificial Intelligence
H. Moosaei, S. Ketabchi, M. Razzaghi, M. Tanveer
Summary: In this paper, two efficient approaches of twin support vector machines (TWSVM) are proposed, including reformulating the formulation by introducing different norms and presenting an efficient algorithm using the generalized Newton's method. Experimental results demonstrate that the new methods outperform baseline methods in terms of performance and computational time.
NEURAL PROCESSING LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Jiamin Xu, Huamin Wang, Libo Zhang, Shiping Wen
Summary: In this paper, a twin depth support vector machine (TDSVM) is proposed, which considers the influence of depth when calculating the distance. By strengthening the center and weakening the edge, a robust SVM framework is constructed, which can identify outliers in the dataset and achieve better generalization performance.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Industrial
Nick Pepper, Luis Crespo, Francesco Montomoli
Summary: This work demonstrates how to approximate the failure probability of an expensive computational model with reliability requirements using Support Vector Machines. An algorithm is proposed to select informative parameter points to improve the approximation accuracy iteratively. Additionally, a method is provided to quantify the uncertainty in the Limit State Function and estimate an upper bound to the failure probability using geometrical arguments.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Computer Science, Artificial Intelligence
Rongda Chen, Zhixia Yang, Junyou Ye
Summary: This article discusses the challenges of using support vector machine (SVM) models in multiview learning and proposes two multiview classifiers, C-MKNSVM and ?-MKNSVM, which overcome the difficulties by using kernel-free techniques. Experimental results show that these classifiers outperform traditional MVL classifiers.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Zhizheng Liang, Lei Zhang
Summary: In this paper, novel twin support vector machines (TSVMs) are proposed to handle uncertain data, where each uncertain sample is modeled as a random vector with Gaussian distributions. By deriving an important theorem to simplify the models and using a quasi-Newton optimization algorithm, the optimization problem becomes tractable. Experimental results show that the proposed models outperform some existing algorithms in terms of classification performance, especially for uncertain cross-plane problems.
PATTERN RECOGNITION
(2022)
Article
Automation & Control Systems
Pedro Ribeiro Mendes Junior, Terrance E. Boult, Jacques Wainer, Anderson Rocha
Summary: When dealing with real-world recognition problems, it is often necessary to have classification methods that can handle unknown classes and reject samples not seen during training. Existing classifiers are mainly designed for closed-set scenarios, where all test samples are assumed to belong to known classes. However, in open-set scenarios, a test sample may not belong to any known class and must be properly rejected as unknown.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Engineering, Multidisciplinary
Juliana Helena Daroz Gaudencio, Fabricio Alves de Almeida, Joao Batista Turrioni, Roberto da Costa Quinino, Pedro Paulo Balestrassi, Anderson Paulo de Paiva
PRECISION ENGINEERING-JOURNAL OF THE INTERNATIONAL SOCIETIES FOR PRECISION ENGINEERING AND NANOTECHNOLOGY
(2019)
Article
Computer Science, Artificial Intelligence
Fabricio Alves de Almeida, Rodrigo Reis Leite, Guilherme Ferreira Gomes, Jose Henrique de Freitas Gomes, Anderson Paulo de Paiva
DECISION SUPPORT SYSTEMS
(2020)
Article
Computer Science, Information Systems
Dayan Adionel Guimaraes, Edielson Prevato Frigieri, Lucas Jun Sakai
Article
Green & Sustainable Science & Technology
Lucas Guedes de Oliveira, Giancarlo Aquila, Pedro Paulo Balestrassi, Anderson Paulo de Paiva, Anderson Rodrigo de Queiroz, Edson de Oliveira Pamplona, Ulisses Pessin Camatta
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2020)
Article
Engineering, Electrical & Electronic
Fabricio Alves de Almeida, Jacques Miranda Filho, Leandro Framil Amorim, Jose Henrique de Freitas Gomes, Anderson Paulo de Paiva
ELECTRIC POWER SYSTEMS RESEARCH
(2020)
Article
Automation & Control Systems
Luiz Gustavo Paes de Souza, Jose Edmilson Martins Gomes, Etory Madrilles Arruda, Gilbert Silva, Anderson Paulo de Paiva, Joao Roberto Ferreira
Summary: This study proposes a robust optimization strategy for roughness optimization in hard turning, considering the trade-off between cutting time and roughness targets along with the variation caused by flank wear. Experimental results demonstrate the effectiveness of the proposed strategy and the ability to explore different solutions to meet varied specifications, preventing the production of non-conforming parts and underutilization of cutting tools.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Fabricio Alves De Almeida, Luiz Gustavo De Mello, Estevao Luiz Romao, Guilherme Ferreira Gomes, Jose Henrique De Freitas Gomes, Anderson Paulo De Paiva, Jacques Miranda Filho, Pedro Paulo Balestrassi
Summary: This study proposes a new methodology to verify the consistency and sensitivity of linkage methods in cluster formation for voltage sag studies; real data is used, and four different disturbed scenarios are evaluated; Ward method shows 100% consistency, making it the most robust method.
Article
Engineering, Electrical & Electronic
Lucas S. Costa, Dayan A. Guimaraes, Edielson P. Frigieri, Rausley A. A. de Souza
Summary: The WCFCPSC algorithm is an improved version of the CFCPSC, enhancing performance while maintaining low complexity and robustness against dynamical noise. Theoretical findings are supported by simulation results, demonstrating the superior performance of WCFCPSC.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Multidisciplinary Sciences
Leandro Duarte Pereira, Pedro Paulo Balestrassi, Vinicius de Carvalho Paes, Anderson Paulo de Paiva, Rogerio Santana Peruchi, Rona Rinston Amauri Mendes
ACTA SCIENTIARUM-TECHNOLOGY
(2020)
Article
Computer Science, Information Systems
Rogerio Santana Peruchi, Paulo Rotela Junior, Tarcisio G. Brito, Anderson P. Paiva, Pedro P. Balestrassi, Lavinia M. Mendes Araujo
Article
Computer Science, Information Systems
Alexandre F. Torres, Franco B. Rocha, Fabricio A. Almeida, Jose H. F. Gomes, Anderson P. Paiva, Pedro Paulo Balestrassi
Article
Computer Science, Information Systems
Fabricio Alves De Almeida, Simone Carneiro Streitenberger, Alexandre Fonseca Torres, Anderson Paulo De Paiva, Jose Henrique De Freitas Gomes
Article
Acoustics
Cailiang Zhang, Zhihui Lai, Zhisheng Tu, Hanqiu Liu, Yong Chen, Ronghua Zhu
Summary: This paper proposes two single-parameter-adjusting SR models to optimize the output performance of SR systems. The effects of the proposed models on SR output under different parameters and signals are investigated through numerical simulations, and their feasibility is verified through experimental results. The research results are of great significance for guiding the design of tri-stable SR models and the application of SR-based signal processing in the context of big data.
Article
Acoustics
Shaoqiong Yang, Hao Chang, Yanhui Wang, Ming Yang, Tongshuai Sun
Summary: In this study, a suspension system based on phononic crystals is designed for vibration isolation of acoustic loads in underwater gliders. The vibration properties of the phononic crystals and the effects of physical parameters on the underwater attenuation zones are investigated. Vibration tests show that the phononic crystal suspension system has a stable vibration isolation effect in the frequency range of 120-5000 Hz.
Article
Acoustics
Xuebin Zhang, Jun Zhang, Tao Liu, Ning Hu
Summary: This study proposes a tunable metamaterial beam to isolate flexural waves. A genetic algorithm-based size optimization is used to obtain a broad low-frequency bandgap. The tunability of the beam is achieved by attaching different numbers of permanent magnets to change the mass of the resonators. Additionally, ultra-broadband flexural wave attenuation is achieved by forming a gradient metamaterial beam based on the rainbow effect. Numerical and experimental results confirm the good flexural wave attenuation ability of the proposed beam.
Article
Acoustics
Luca Rapino, Francesco Ripamonti, Samanta Dallasta, Simone Baro, Roberto Corradi
Summary: This paper presents a method for simulating tyre/road noise using equivalent monopoles, including the synthesis of monopoles through an inverse problem approach and the use of an ISO 10844 road replica for laboratory testing. The method combines acoustic finite element models and numerical simulations of vehicles, and the results are validated by comparing them with measured data.
Article
Acoustics
Xiaoyan Zhu, Tin Oberman, Francesco Aletta
Summary: This paper explores the definition of acoustical heritage and proposes a multidimensional definition based on interviews with experts and detailed analysis of the data.
Article
Acoustics
Faeez Masurkar, Saurabh Aggarwal, Zi Wen Tham, Lei Zhang, Feng Yang, Fangsen Cui
Summary: This research focuses on estimating the elastic constants of orthotropic laminates using ultrasonic guided waves and inverse machine learning models. The results show that this approach has the potential to accurately predict the elastic constants of a material and reduce computational time.
Article
Acoustics
Feng Xiao, Haiquan Liu, Jia Lu
Summary: Diagnostic methods for cardiovascular disease based on heart sound classification have been widely studied due to their noninvasiveness, low-cost, and high efficiency. However, existing research often faces challenges such as the nonstationarity and complexity of heart sound signals, leading to limited capability of neural networks to extract discriminative features. To address these issues, this study proposes a novel convolutional neural network that combines 1D convolution and 2D convolution, and introduces an attention mechanism to enhance feature extraction capability. The study also explores the advantages and disadvantages of combining deep learning features with manual features, and adopts an evolving fuzzy system for decision-making interpretability.
Article
Acoustics
Hong Xu, Zhengyao He, Qiang Shi, Yushi Wang, Bo Zhang
Summary: This paper presents the development of a directional segmented ring transmitting transducer that can radiate sound waves in any horizontal region. The study focuses on the structure of the segmented ring transducer, its radiation sound field characteristics, and the beam pattern control method based on modal synthesis. The authors propose orthogonal beam pattern functions for adjusting steering angles and establish a three-dimensional finite element model to simulate the transmitting beam patterns. Experimental measurements and tests validate the effectiveness of the proposed transducer, showcasing its ability to steer the beam patterns to different directions.
Article
Acoustics
Jirui Yang, Shefeng Yan, Di Zeng, Gang Tan
Summary: This paper proposes an improved domain adaptation framework, self-supervised learning minimax entropy, to enhance the recognition performance of underwater target recognition models. The experimental results demonstrate that applying domain adaptation methods can effectively improve the recognition accuracy of the models under various marine conditions.
Article
Acoustics
Zonghan Sun, Jie Tian, Yuhang Zheng, Xiaocheng Zhu, Zhaohui Du, Hua Ouyang
Summary: This paper analyzes the noise reduction method of installing a sinusoidal-shaped inlet duct on a cooling fan through theoretical and experimental analysis of the acoustic mode modulation. The study establishes the correlation between the free field noise and acoustic mode of the fan rotor and the unsteady forces on the rotor blade surface. The results show that the sinusoidal-shaped inlet duct achieves greater noise reduction compared to a straight duct, especially at the blade passing frequency and its first harmonic.
Article
Acoustics
Min Li, Rumei Han, Hui Xie, Ruining Zhang, Haochen Guo, Yuan Zhang, Jian Kang
Summary: This study is part of a global collaboration to translate and standardise soundscape research. A reliable questionnaire for soundscape characterisation in Mandarin Chinese was developed and validated. The study found that salient sound sources become the focus of attention for individuals in urban open spaces, and the perception is also influenced by the acoustic characteristics of the soundscape. Certain types of sound sources play a more important role in soundscape perception.
Article
Acoustics
Arezoo Talebzadeh, Dick Botteldooren, Timothy Van Renterghem, Pieter Thomas, Dominique Van de Velde, Patricia De Vriendt, Tara Vander Mynsbrugge, Yuanbo Hou, Paul Devos
Summary: This study proposes a sound selection methodology to enhance the soundscape in nursing homes and reduce BPSD by analyzing sound characteristics and recognition methods. The results highlight the sound characteristics that lead to positive responses, while also pointing out the need for further studies to understand which sounds are most suitable for people with dementia.
Article
Acoustics
Yang Yang, Yongxin Yang, Zhigang Chu
Summary: This paper introduces a grid-free compressive beamforming method compatible with arbitrary linear microphone arrays, and demonstrates the correctness and superiority of the proposed method through examples. Monte Carlo simulations are performed to reveal the effects of source coherence, source separation, noise, and number of snapshots.
Article
Acoustics
Sukru Selim Calik, Ayhan Kucukmanisa, Zeynep Hilal Kilimci
Summary: Computer-Aided Language Learning (CALL) is growing rapidly due to the importance of acquiring proficiency in multiple languages for effective communication. In the field of CALL, the detection of mispronunciations is vital for non-native speakers. This research introduces a novel framework using audio-centric transformer models to detect mispronunciations in Arabic phonemes. The results demonstrate that the UNI-SPEECH transformer model yields notable classification outcomes in Arabic phoneme mispronunciation detection. The comprehensive comparison of these transformer models provides valuable insights and guidance for future investigations in this domain.
Article
Acoustics
Yi-Yang Ni, Fei-Yun Wu, Hui-Zhong Yang, Kunde Yang
Summary: This paper proposes an improved method for compressive sensing by introducing a self training dictionary scheme and a CS reconstruction method based on A*OLS, which enhances the sparse representation performance of propeller signals.
