Article
Computer Science, Information Systems
Chaoyu Gong, Zhi-gang Su, Xinyi Zhang, Yang You
Summary: This paper presents an adaptive evidential K-NN classification (AEK-NN) algorithm that addresses several issues in the K-NN algorithm, including the potential for incorrect classification results due to the use of a predetermined K, and the curse of dimensionality in high-dimensional spaces. By incorporating adaptive neighborhood search and feature weighting, the algorithm is able to effectively handle data with imperfect labels.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Antonio Javier Gallego, Juan Ramon Rico-Juan, Jose J. Valero-Mas
Summary: The paper introduces the caKD+ algorithm which combines various techniques to improve the efficiency of kNN search, outperforming 16 state-of-the-art methods on 10 datasets.
PATTERN RECOGNITION
(2022)
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
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
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
Polychronis Velentzas, Michael Vassilakopoulos, Antonio Corral
Summary: Edge computing aims to improve performance by storing and processing data closer to their source. This paper proposes a distributed edge-computing environment architecture for large-scale processing of the k-NN query using an efficient algorithm on GPU and SSD-enabled edge nodes. The new algorithm outperforms existing ones in experimental performance evaluations.
INTERNET OF THINGS
(2021)
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
Chemistry, Multidisciplinary
Cuixia Li, Shanshan Yang, Li Shi, Yue Liu, Yinghao Li
Summary: This paper proposes an end-to-end point cloud registration network model called PTRNet, which improves the registration behavior by considering both local and global features. Experimental results show that PTRNet outperforms other methods in terms of average error and registration accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Naz Gul, Muhammad Aamir, Saeed Aldahmani, Zardad Khan
Summary: The paper proposes a $k$ NN ensemble method that identifies nearest observations based on weighted distances using support vectors, showing better classification performance on datasets with noisy features. Through majority voting, the estimated class of a test observation is decided, outperforming other methods in most cases.
Article
Computer Science, Artificial Intelligence
P. Jayapriya, K. Umamaheswari
Summary: This research proposes an effective feature optimization technique using the K-nearest neighbor algorithm and differential evolution for finger knuckle print-based authentication. Experimental results show improved classification accuracy with this method.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Namrata Karlupia, Pawanesh Abrol
Summary: Nature-inspired computing, which mimics natural processes, provides machine solutions to complex problems. The challenge of high-dimensional data with redundant features is addressed using metaheuristic techniques, particularly the whale optimization algorithm (WOA). In this study, five nature-inspired algorithms were compared for feature selection, and WOA was found to perform the best.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Ahmed Ezzat, Basem E. Elnaghi, Abdelazeem A. Abdelsalam
Summary: This paper proposes a methodology for identifying the decision of microgrid islanding through feature extraction and K-nearest neighbor technique. The methodology shows high accuracy and short detection time in determining islanding decisions even in the presence of noise, with the ability to effectively distinguish between islanding and non-islanding.
ELECTRIC POWER SYSTEMS RESEARCH
(2021)
Article
Computer Science, Interdisciplinary Applications
Ping Lu, Shiyuan Guo, Yang Shu, Bin Liu, Peifeng Li, Wei Cao, Kaiyong Jiang
Summary: Natural element method (NEM) is a meshless method that simplifies the imposition of essential boundary conditions and has great potential to solve problems with large deformation. However, the high computational cost for searching natural neighbors is a main challenge in NEM. A local algorithm based on K-Nearest Neighbor is proposed in this paper to reduce the search scope and improve efficiency.
ADVANCES IN ENGINEERING SOFTWARE
(2023)
Article
Biology
Omneya Attallah, Shaza Zaghlool
Summary: Pediatric medulloblastomas (MBs), the most common malignant brain tumors in children, are heterogenous and challenging to classify accurately based on histopathological images. This study combines textural analysis and deep learning techniques to improve the subtype identification of pediatric MBs. The automated pipeline proposed in this study shows an increased accuracy in classification compared to previous methods, providing a powerful tool for individualized therapies and identification of high-risk complications in children.
Article
Engineering, Biomedical
Omneya Attallah, Dina A. Ragab
Summary: This paper proposes an automated diagnostic tool, Auto-MyIn, for diagnosing myocardial infarction (MI) using multiple convolutional neural networks (CNN). The tool utilizes textural information obtained from grey level co-occurrence matrix (GLCM) and principal component analysis (PCA) to improve diagnostic accuracy. The results show that fusing textural-based deep features and using textural information is superior to using spatial information of the original DE-MRI images. The performance of Auto-MyIn indicates its reliability and competitive ability compared to other related studies.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Automation & Control Systems
Omneya Attallah
Summary: An automated tool called RADIC was developed to train deep learning models using radiomics-generated images, improving the accuracy of COVID-19 diagnosis with a final accuracy of 99.4% and 99% on two benchmark datasets.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Omneya Attallah
Summary: Routine checks can help prevent cervical cancer. Artificial intelligence can enhance the traditional testing procedure and improve diagnostic accuracy.
APPLIED SCIENCES-BASEL
(2023)
Article
Medicine, General & Internal
Omneya Attallah
Summary: Retinopathy of prematurity (ROP) is a serious ocular problem in premature infants. This paper proposes an automated CAD tool called GabROP, which uses Gabor wavelets and multiple deep learning models for ROP diagnosis. GabROP analyzes fundus images using Gabor wavelets and trains three CNN models independently. The features from these models are fused using discrete cosine transform (DCT) to obtain the final diagnosis result. Experimental results demonstrate the accuracy and efficiency of GabROP.
Article
Engineering, Multidisciplinary
Ahmed H. Salama, Dina A. Ragab, Nancy M. Abdel-Moneim
Summary: During the Covid-19 pandemic, the importance of outdoor public spaces for maintaining people's mental health is highlighted. An online survey, statistical analysis, and machine learning techniques were used to identify the significance of outdoor spaces and categorize user preferences. The results showed that 85.17% of the sample acknowledged the importance of outdoor public spaces. However, further research is needed to reconsider urban design and relocate indoor activities outdoors to preserve human mental well-being.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Omneya Attallah
Summary: This study proposes an efficient computer-aided diagnostic (CAD) system called CerCan.Net for the automatic diagnosis of cervical cancer. It utilizes lightweight convolutional neural networks (CNNs) to accurately and rapidly diagnose cervical cancer, overcoming the limitations of traditional diagnosis.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Chemistry, Analytical
Omneya Attallah
Summary: Given the rise in gas leakage accidents at coal mines, chemical companies, and home appliances, rapid and automated gas detection and identification are necessary. This paper proposes a deep learning-based pipeline that combines multimodal data obtained from infrared thermal imaging and an electronic nose sensor array. The results show a high accuracy of gas detection using the proposed multimodal fusion approach.
Article
Engineering, Multidisciplinary
Omneya Attallah
Summary: A new method using deep learning techniques is proposed to identify and categorize rice paddy diseases. It improves recognition accuracy by collecting features from different layers and applying spectral-temporal information, and reduces recognition complexity through feature selection.
Article
Health Care Sciences & Services
Omneya Attallah
Summary: In this study, a computer-aided diagnosis tool named Monkey-CAD was developed to rapidly and accurately diagnose monkeypox. The tool extracted features from eight convolutional neural networks and explored the best combination of these features to influence classification. The results showed that Monkey-CAD could assist healthcare practitioners in diagnosis and fusion of deep features from selected CNNs could improve performance.
Article
Horticulture
Omneya Attallah
Summary: Tomatoes are valuable vegetables and an economic pillar for many countries. Automatic identification of tomato leaf diseases using deep learning models has been widely studied, but existing methods suffer from high computational complexity and large dimensionality. This study proposes a pipeline that utilizes compact convolutional neural networks and transfer learning for condensed and high-level representation. The pipeline also applies a hybrid feature selection approach to reduce dimensions and achieves high accuracy in tomato leaf disease identification.
Proceedings Paper
Computer Science, Interdisciplinary Applications
Dawlat Al-Saadany, Omneya Attallah, Khaled Elzaafarany, A. A. A. Nasser
Summary: Fetal arrhythmia, caused by a problem in the fetus's heart's electrical system, is an abnormal heart rhythm that requires monitoring for providing valuable information about the fetus's condition. Current methods involving multiple electrodes for acquiring abdominal ECG from the mother cause discomfort and difficulty in extracting ECG due to noise and artifacts. This study presents a machine learning framework for detecting fetal arrhythmia using a single abdominal ECG, achieving a high accuracy rate.
COMPUTATIONAL SCIENCE, ICCS 2022, PT II
(2022)
Article
Health Care Sciences & Services
Omneya Attallah
Summary: With the current health crisis caused by the COVID-19 pandemic, patients' preference for limited contact with doctors or clinicians has led to the development of computer-aided facial diagnosis systems. This study introduces FaceDisNet, a novel system that utilizes deep learning techniques and a new public dataset to diagnose single and multiple diseases accurately without physical contact with patients. The high accuracy achieved by FaceDisNet demonstrates its reliability and potential for assisting physicians in manual diagnosis.
Proceedings Paper
Computer Science, Artificial Intelligence
Ossama Rashad, Omneya Attallah, Iman Morsi
Summary: This paper proposes a PLC/HMI system for tracking oil products refineries, which utilizes AOI programming and PLC to automatically display petroleum products terminal. The results of the study show that the system can simplify the program, decrease scan time, and lower system creation and upgrade costs.
5TH INTERNATIONAL CONFERENCE ON COMPUTING AND INFORMATICS (ICCI 2022)
(2022)
Article
Health Care Sciences & Services
Omneya Attallah
Summary: The accurate and rapid detection of novel coronavirus is crucial to prevent its spread, and artificial intelligence techniques can aid in this process. This study proposes a computer-assisted diagnostic framework based on deep learning and texture-based radiomics approaches. By fusing deep features from multiple convolutional neural networks, the diagnostic accuracy is improved. The performance of this framework allows radiologists to achieve fast and accurate diagnosis of coronavirus.