Article
Physics, Multidisciplinary
Congcong Feng, Bo Zhao, Xin Zhou, Xiaodong Ding, Zheng Shan
Summary: The K-nearest neighbor (KNN) algorithm is widely used in classification, but its high time complexity affects its performance in big data era. The quantum K-nearest neighbor (QKNN) algorithm can handle this problem with high efficiency, but its accuracy is compromised when using the traditional similarity measure based on Euclidean distance. This work proposes a new similarity measure, Polar distance, inspired by the Polar coordinate system and quantum property, which considers both angular and module length information and introduces a weight parameter adjusted to the specific application data.
Article
Computer Science, Hardware & Architecture
Martin Aumueller, Sariel Har-Peled, Sepideh Mahabadi, Rasmus Pagh, Francesco Silvestri
Summary: This paper studies the r-NN problem in similarity search in the context of individual fairness and equal opportunities. The authors propose efficient data structures for the fair NN problem and highlight the inherent unfairness of existing NN data structures through experimental evaluation.
COMMUNICATIONS OF THE ACM
(2022)
Article
Automation & Control Systems
Hongjiao Guan, Long Zhao, Xiangjun Dong, Chuan Chen
Summary: Imbalanced data classification is a challenging problem in many applications. We propose an extended natural neighbor (ENaN) concept without parameter k to improve the quality of generated examples by accurately reflecting the local distribution. ENaN-based SMOTE (ENaNSMOTE) can improve the sample distribution obtained by SMOTE and NaNSMOTE.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Chemistry, Multidisciplinary
Jianfeng Xi, Shiqing Wang, Tongqiang Ding, Jian Tian, Hui Shao, Xinning Miao
Summary: The study collected real driving data to establish a fatigue driving behavior detection model, and compared the models based on different algorithms, finding that the model based on the k-nearest neighbor algorithm is more reliable.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Jianping Gou, Liyuan Sun, Lan Du, Hongxing Ma, Taisong Xiong, Weihua Ou, Yongzhao Zhan
Summary: This article proposes a novel representation coefficient-based k-nearest centroid neighbor method (RCKNCN) aiming to improve the classification performance and reduce the sensitivity to the neighborhood size k. The method captures both the proximity and geometry of k-nearest neighbors and learns to differentiate the contribution of each neighbor to the classification of a testing sample. A weighted majority voting algorithm is also proposed under the RCKNCN framework.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Yuming Wu, Lei Zhang, Ren Lou, Xinghua Li
Summary: The increasing number of vehicles has made traffic safety more complicated. Autonomous vehicles (AVs) have the potential to greatly reduce accidents. This study proposes a lane changing maneuver recognition model based on physical data and machine learning, achieving good results.
Article
Computer Science, Artificial Intelligence
Benqiang Wang, Shunxiang Zhang
Summary: The study proposes a new locally adaptive k-nearest centroid neighbour classification method based on average distance, which improves classification performance by finding nearest centroid neighbours to determine k neighbours and deriving discrimination classes with different k values based on the number and distribution of neighbours, resulting in better performance compared to other state-of-the-art KNN algorithms.
CONNECTION SCIENCE
(2022)
Article
Computer Science, Information Systems
Wan-Lei Zhao, Hui Wang, Peng-Cheng Lin, Chong-Wah Ngo
Summary: This paper addresses the issue of merging k-nearest neighbor (k-NN) graphs in two different scenarios. A symmetric merge algorithm is proposed to combine two approximate k-NN graphs, facilitating large-scale processing. A joint merge algorithm is also proposed to expand an existing k-NN graph with a raw dataset, enabling the incremental construction of a hierarchical approximate k-NN graph.
IEEE TRANSACTIONS ON BIG DATA
(2022)
Article
Chemistry, Multidisciplinary
Dun Hu, Lifu Gao
Summary: The study proposed an HRV analysis method based on heartbeat modes to detect drivers' stress, utilizing statistical linguistic tools to quantify the heart rate time series and identify different heartbeat modes. Using the k-nearest neighbors (k-NN) classification algorithm, the probability of each heartbeat mode was employed as a feature to detect and recognize stress caused by the driving environment, with an accuracy of 93.7%.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Yibang Ruan, Yanshan Xiao, Zhifeng Hao, Bo Liu
Summary: The paper introduces a nearest-neighbor search model for distance metric learning (NNS-DML), which constructs metric optimization constraints by searching different optimal nearest-neighbor numbers for each training instance. This model reduces the influence of irrelevant features on similar and dissimilar instance pairs and develops a k-free nearest-neighbor model for classification problems. Extensive experiments show that NNS-DML outperforms state-of-the-art distance metric learning methods.
INFORMATION SCIENCES
(2021)
Article
Quantum Science & Technology
Jing Li, Song Lin, Kai Yu, Gongde Guo
Summary: This paper proposes a quantum K-nearest neighbor classification algorithm based on the Hamming distance, which utilizes quantum computation to obtain the Hamming distance in parallel and combines a core sub-algorithm for finding the minimum distance. The whole quantum frame of the K-nearest neighbor classification algorithm is presented, and the proposed algorithm achieves a significant speedup.
QUANTUM INFORMATION PROCESSING
(2022)
Article
Food Science & Technology
Raira Sa de Brito, Marcos Jhony Almeida Costa, Jhonatas Rodrigues Barbosa, Adilson Ferreira Santos Filho, Fabricio de Souza Farias, Lucia de Fatima Henriques Lourenco
Summary: Nanoparticles associated with antioxidants show promise in improving food packaging. Gelatin-carboxymethylcellulose films containing propolis extract and TiO2 nanoparticles were developed and the K-Nearest Neighbor algorithm was used to select the best composition. Films with nanoparticles demonstrated superior antioxidant and barrier properties. The film with propolis and 5% TiO2 nanoparticles showed the highest antioxidant activity and desirable physical properties for food packaging. The KNN algorithm proves to be a useful tool for classifying and selecting films with desired properties.
FOOD PACKAGING AND SHELF LIFE
(2023)
Article
Chemistry, Analytical
Dmytro Chumachenko, Mykola Butkevych, Daniel Lode, Marcus Frohme, Kurt J. G. Schmailzl, Alina Nechyporenko
Summary: Diagnosis of cardiovascular diseases is urgent, and machine learning methods have shown high accuracy in classifying patients with suspected myocardial infarction.
Article
Computer Science, Artificial Intelligence
Sanjoy Chakraborty, Apu Kumar Saha, Absalom E. Ezugwu, Ratul Chakraborty, Ashim Saha
Summary: The study presents an enhanced version of the Whale Optimization Algorithm (WOA) called HCCWOA, which incorporates horizontal crossover, co-operative learning techniques, and inertia weight to improve the exploration and exploitation capabilities. The effectiveness of HCCWOA is evaluated on twelve datasets and compared with other algorithms, demonstrating its superior performance. Statistical analyses further support the efficacy of HCCWOA in effectively exploring feature spaces and selecting relevant characteristics for classification tasks.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Biology
Qian Zhang, Jinhua Sheng, Qiao Zhang, Luyun Wang, Ze Yang, Yu Xin
Summary: To diagnose Alzheimer's disease and its early stage, mild cognitive impairment, a framework utilizing magnetic resonance imaging (MRI) image preprocessing, feature extraction, and the FKNN model is proposed in this paper. The framework incorporates an improved Harris Hawks Optimization algorithm called SSFSHHO, which employs the Sobol sequence and Stochastic Fractal Search mechanisms to optimize the parameters of FKNN. Experimental results demonstrate that the SSFSHHO-FKNN model achieves high classification performance for AD and MCI, outperforming other comparative algorithms.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Engineering, Biomedical
Shun-Min Samuel Wang, Yi-Jing Huang, Jia-Jin Jason Chen, Chun-Wei Wu, Chien-An Chen, Che-Wei Lin, Van-Truong Nguyen, Chih-Wei Peng
Summary: A novel HD transcranial burst electrostimulation device was designed and tested for its therapeutic potential in neurorehabilitation. The safety and accuracy of the device were validated through a series of in vitro experiments. A pilot clinical trial demonstrated that the active HD transcranial burst electrostimulation group showed greater improvement in voluntary motor function and coordination of the upper extremity compared to the sham control group, with no severe adverse events noted.
JOURNAL OF NEURAL ENGINEERING
(2021)
Article
Chemistry, Analytical
Febryan Setiawan, Che-Wei Lin
Summary: This study developed a novel detection and severity classification algorithm using deep learning approaches to classify Parkinson's disease severity based on vertical ground reaction force (vGRF) signals. By utilizing techniques such as continuous wavelet transform and principal component analysis, the algorithm achieved a high average accuracy of 96.52% using ResNet-50 as a classifier for classification. The algorithm aims to assist physicians in early detection, effective treatment planning, and disease progression monitoring for PD patients.
Article
Neurosciences
Febryan Setiawan, Che-Wei Lin
Summary: A novel identification algorithm using deep learning was developed to classify neurodegenerative diseases based on vertical ground reaction force signals. The algorithm, which includes preprocessing, feature transformation, and classification processes, effectively differentiated gait patterns between healthy controls and NDD patients.
Article
Geriatrics & Gerontology
Guan-Bo Chen, Che-Wei Lin, Hung-Ya Huang, Yi-Jhen Wu, Hung-Tzu Su, Shu-Fen Sun, Sheng-Hui Tuan
Summary: Virtual reality-based progressive resistance training was shown to be partially effective in improving handgrip strength and walking speed among older sarcopenic adults in health care facilities.
JOURNAL OF AGING AND PHYSICAL ACTIVITY
(2021)
Article
Medicine, General & Internal
Cheng-Yu Lin, Yi-Wen Wang, Febryan Setiawan, Nguyen Thi Hoang Trang, Che-Wei Lin
Summary: This study proposes an algorithm using machine learning and ECG spectrogram-derived bag-of-features for detecting sleep apnea. The algorithm was validated using ECG recordings from 83 subjects and achieved high accuracy and temporal resolution.
JOURNAL OF CLINICAL MEDICINE
(2022)
Article
Chemistry, Analytical
Ke-Wei Chen, Laura Bear, Che-Wei Lin
Summary: This study improves on the current solution methods for electrocardiographic imaging (ECGi) using machine learning and deep learning frameworks. By simultaneously recording electrocardiograms from pigs' ventricles and body surfaces, and constructing a model using neural networks, the study finds that relatively small datasets can achieve accuracy compatible with current standard methods.
Article
Biology
Febryan Setiawan, Che-Wei Lin
Summary: In this study, a novel and robust sleep apnea (SA) detection algorithm based on a deep learning framework was developed. The algorithm preprocessed and decomposed the ECG signal, and achieved good classification performance for normal and apnea events.
Article
Chemistry, Analytical
Hsiao-Ting Fu, Hui-Zin Tu, Herng-Sheng Lee, Yusen Eason Lin, Che-Wei Lin
Summary: This study evaluated the performance of an AI-based TB automated system that uses a microscopic scanner and recognition program to detect and classify acid-fast bacilli. The system showed significant improvements in accuracy, sensitivity, and specificity after two stages of testing, and demonstrated enhanced laboratory efficiency.
Article
Engineering, Biomedical
Sheng-Chiao Lin, Ming-Yee Lin, Bor-Hwang Kang, Yaoh-Shiang Lin, Yu-Hsi Liu, Chi-Yuan Yin, Po-Shing Lin, Che-Wei Lin
Summary: This study aimed to examine the effect of coherent frequency on hearing prognosis in patients with SSNHLV undergoing high-dose steroid treatment using vHIT. A retrospective cohort study was conducted, and 64 patients were included in the analysis. The results showed that a higher coherent frequency in the posterior SCC was associated with complete recovery of hearing. Furthermore, the use of CNN and SVM in WCA classification and feature extraction improved accuracy and stability in predicting treatment outcomes. In conclusion, high coherent frequency in vHIT can lead to favorable hearing outcomes in SSNHLV and aid in AI classification.
IEEE JOURNAL OF TRANSLATIONAL ENGINEERING IN HEALTH AND MEDICINE
(2023)
Article
Health Care Sciences & Services
Huey-Pin Tsai, Che-Wei Lin, Ying-Jun Lin, Chun-Sheng Yeh, Yan-Shen Shan
Summary: A VR software was developed and implemented to simulate high-level virological testing, which proved to meet the participants' needs and enhance their interest in learning. The use of VR training significantly improved participants' posttraining scores and knowledge of specific items. The VR program can reduce training costs, increase accessibility, and enhance practical skills and learning motivation.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Article
Mathematics
You-Liang Xie, Che-Wei Lin
Summary: This study presents a low-memory-usage ectopic beat classification convolutional neural network (CNN) (LMUEBCNet) and a correlation-based oversampling (Corr-OS) method for ectopic beat data augmentation. The LMUEBCNet classifier consists of four VGG-based convolution layers and two fully connected layers with the continuous wavelet transform (CWT) spectrogram as input. The Corr-OS method augments synthetic beats using the top K correlation heartbeats for balancing the training set.
Article
Medicine, General & Internal
Wei-Min Liu, Che-Lun Yeh, Po-Wei Chen, Che-Wei Lin, An-Bang Liu
Summary: Simple algorithms utilizing keystroke biometric parameters can effectively distinguish de novo PD patients. Typing speed and number of words typed were identified as reliable tools for assessing clinical statuses in PD patients.
Article
Computer Science, Information Systems
Febryan Setiawan, An-Bang Liu, Che-Wei Lin
Summary: A detection algorithm for neurodegenerative disease (NDD) was developed using a convolutional neural network (CNN) and wavelet coherence spectrogram. The algorithm effectively classifies NDD based on gait force signals, aiding physicians in screening, early diagnosis, treatment planning, and disease progression monitoring.
Article
Clinical Neurology
Hsiu-Yun Hsu, Li-Chieh Kuo, Yu-Ching Lin, Fong-Chin Su, Tai-Hua Yang, Che-Wei Lin
Summary: The study found that embedding mirror therapy within a virtual reality system may have potential effects on restoring sensorimotor function of the upper limb in chronic stroke patients. However, there is rather weak evidence regarding the additive effect of virtual reality on improving the primary outcome, and further research is needed.
NEUROREHABILITATION AND NEURAL REPAIR
(2022)
Article
Computer Science, Information Systems
Che-Wei Lin, Li-Chieh Kuo, Yu-Ching Lin, Fong-Chin Su, Yu-An Lin, Hsiu-Yun Hsu
Summary: The study found that the virtual reality mirror therapy system has a positive impact on the sensorimotor performance of hands in healthy participants, and also provides beneficial effects on motor function of the upper extremity in chronic stroke patients.
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.