Article
Computer Science, Artificial Intelligence
Shili Peng, Wenwu Wang, Yinli Chen, Xueling Zhong, Qinghua Hu
Summary: This article presents a new idea for addressing the challenge of unifying classification and regression in machine learning. It proposes converting the classification problem into a regression problem and using regression methods to solve key problems in classification. Experimental results demonstrate that the proposed method outperforms existing algorithms in terms of prediction accuracy and model uncertainty.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Dhruba Jyoti Kalita, Vibhav Prakash Singh, Vinay Kumar
Summary: Breast cancer is a common cause of death among women globally. A novel 2-way threshold-based Intelligent Water Drops (IWD) algorithm for feature selection has been proposed in this study to design an effective and efficient Computer-Aided Diagnosis (CAD) system for early-stage detection of breast cancer. The approach shows superior performance compared to other meta-heuristic feature selection techniques, achieving high accuracy and precision rates in classifying new mammograms.
Article
Computer Science, Artificial Intelligence
Dounia Lakhmiri, Ryan Alimo, Sebastien Le Digabel
Summary: This study introduces Delta-MADS, a hybrid derivative-free optimization method, designed to quickly produce efficient variational autoencoders to assist the GDSA team in their mission.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Chemistry, Multidisciplinary
Ningyuan Zhu, Chaoyang Zhu, Liang Zhou, Yayun Zhu, Xiaojuan Zhang
Summary: This paper introduces an intrusion detection method for power industrial control systems and proposes an improved algorithm to optimize the performance of the model. The experimental results demonstrate that the method achieves superior detection performance and outperforms similar approaches.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
H. S. Laxmisagar, M. C. Hanumantharaju
Summary: This paper introduces the use of Support Vector Machine (SVM) algorithm to solve regression and classification problems. In order to improve performance and reduce cost and power consumption, researchers attempted to implement SVM algorithm on hardware. A hardware-based SVM linear classifier with pipeline architecture is proposed for fast processing using Verilog HDL. The performance metrics of resource utilization, power consumption, and timing are evaluated, and the accuracy rate is compared with software.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Hardware & Architecture
Mehmet Akif Bulbul
Summary: This study focuses on the diagnosis of breast cancer using genetic algorithms and artificial neural networks. A hybrid model is created and compared with other machine learning methods, showing a high accuracy rate. An iOS-android-based application is also developed to aid patients in making decisions.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Chunnan Wang, Hongzhi Wang, Chang Zhou, Hanxiao Chen
Summary: Machine learning models are highly sensitive to hyperparameters and their evaluations can be costly. The ExperienceThinking algorithm proposed in this paper intelligently optimizes hyperparameter settings based on known evaluation information, effectively improving the performance of machine learning models within limited budgets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Luntian Mou, Jiali Chang, Chao Zhou, Yiyuan Zhao, Nan Ma, Baocai Yin, Ramesh Jain, Wen Gao
Summary: Distracted driving is a major cause of traffic accidents, so it is important for intelligent vehicles to establish a distraction detection system that can continuously monitor driver behavior and respond accordingly. Most existing research focuses on either local features or global features, but we propose a novel dual-channel feature extraction model based on CNN and Transformer that integrates both local and global features effectively.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Automation & Control Systems
Quentin Bertrand, Quentin Klopfenstein, Mathurin Massias, Mathieu Blondel, Samuel Vaiter, Alexandre Gramfort, Joseph Salmon
Summary: This paper studies the optimization problem when the internal problem is convex but non-smooth and proposes two methods to accelerate computation. Experimental results show that this optimization method has computational advantages in hyperparameter optimization, especially when multiple hyperparameters are required.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Multidisciplinary Sciences
Qingbo Li, Zhixiang Zhang, Zhenhe Ma
Summary: Raman spectroscopy provides valuable information for early detection and diagnosis of breast cancer by measuring molecular components and structure. The study uses feature fusion strategy to reduce dimensionality and enhance discrimination between normal tissues and tumors. Through random experiments, the classifier achieved over 96% performance in accuracy, sensitivity, and specificity.
Article
Computer Science, Information Systems
Mahmoud Ragab, Maged Mostafa Mahmoud, Amer H. Asseri, Hani Choudhry, Haitham A. Yacoub
Summary: This study introduces a new algorithm called SMADTL-CCDC for colorectal cancer detection and classification, which utilizes deep learning techniques and image processing. The algorithm shows improved performance compared to recent approaches, with innovative methods in pre-processing, feature extraction, and recognition.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Interdisciplinary Applications
Varanasi Satya Sreekanth, Karnam Raghunath, Deepak Mishra
Summary: Atmospheric Gravity Waves play a significant role in Middle Atmosphere Dynamics, and the breaking of Gravity Waves leads to turbulence. However, the accuracy and sparsity of Wind Velocity measuring instruments at the altitude of interest pose a problem for confirming the breaking of Atmospheric Gravity Waves. In this study, we propose a solution using Dictionary Learning and Deep Learning methods to detect Wave Breaking events from atmospheric temperature perturbations, and the effectiveness of this method is demonstrated through a case study using satellite data and validated with ground-based instruments.
COMPUTERS & GEOSCIENCES
(2023)
Article
Computer Science, Information Systems
Yesi Novaria Kunang, Siti Nurmaini, Deris Stiawan, Bhakti Yudho Suprapto
Summary: The research proposes a deep learning intrusion detection system that combines a pretraining approach with deep autoencoder and deep neural network through hyperparameter optimization procedures. Testing on the NSL-KDD and CSE-CIC-ID2018 datasets shows improved performance metrics in multiclass classification.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2021)
Article
Construction & Building Technology
Saman Taheri, Amirhossein Ahmadi, Behnam Mohammadi-Ivatloo, Somayeh Asadi
Summary: Deep learning algorithms, particularly deep recurrent neural networks (DRNNs), have gained attention for fault detection diagnostic in HVAC systems due to their high detection accuracy. Challenges include exploring bespoke DRNN configurations and optimizing hyperparameters, but the study successfully introduces and compares different configurations to achieve high performance. The final DRNN model outperforms other advanced data-driven techniques such as random forest and gradient boosting, showcasing its effectiveness in fault detection for HVAC systems.
ENERGY AND BUILDINGS
(2021)
Article
Computer Science, Information Systems
Yassir Edrees Almalki, Ahmad Shaf, Tariq Ali, Muhammad Aamir, Sharifa Khalid Alduraibi, Shoayea Mohessen Almutiri, Muhammad Irfan, Mohammad Abd Alkhalik Basha, Alaa Khalid Alduraibi, Abdulrahman Manaa Alamri, Muhammad Zeeshan Azam, Khalaf Alshamrani, Hassan A. Alshamrani
Summary: This study utilizes a hybrid model combining convolutional neural network (CNN) and support vector machine (SVM) to successfully classify and recognize breast mammogram images, achieving better performance compared to existing methods.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Physics, Multidisciplinary
Saad Ahmad, Shubham Agrawal, Samta Joshi, Sachin Taran, Varun Bajaj, Fatih Demir, Abdulkadir Sengur
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2020)
Article
Acoustics
Fatih Demir, Muammer Turkoglu, Muzaffer Aslan, Abdulkadir Sengur
Article
Computer Science, Artificial Intelligence
Fatih Demir
Summary: The global spread of the COVID-19 pandemic has caused severe shortages of resources and highlighted the need for deep learning-based applications to assist. A new approach based on a deep LSTM model for automatically identifying COVID-19 cases from X-ray images was proposed, showing promising results through image pre-processing.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Hardware & Architecture
Fatih Demir, Kursat Demir, Abdulkadir Sengur
Summary: In this study, a new method based on a deep learning model was proposed to classify chest X-ray images containing COVID-19, normal, and pneumonia cases. By utilizing a trained model for feature extraction and using an SVM classifier for classification, the proposed method achieved a high accuracy of 99.75% and outperformed other deep learning-based approaches in terms of performance.
NEW GENERATION COMPUTING
(2022)
Article
Health Care Sciences & Services
Fatih Demir, Kamran Siddique, Mohammed Alswaitti, Kursat Demir, Abdulkadir Sengur
Summary: This study proposes a new approach based on multi-level feature selection to detect Parkinson's disease by analyzing voice recordings. Feature selection was performed using the Chi-square, L1-Norm SVM, and ReliefF algorithms, and machine learning was done using the KNN classifier. The proposed approach achieved a classification accuracy of 95.4%.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Health Care Sciences & Services
Fatih Demir, Burak Tasci
Summary: The study proposed a new method for diagnosing ophthalmological diseases by analyzing deep features in fundus images and using the SVM algorithm for classification, accurately detecting eight different ophthalmological diseases. Additionally, a multilevel feature selection algorithm called NCAR was used to enhance the accuracy of detection.
JOURNAL OF PERSONALIZED MEDICINE
(2021)
Article
Engineering, Biomedical
Fatih Demir, Yaman Akbulut
Summary: This study presents a new deep learning-based approach for automatic brain tumor detection using MRI images. The approach achieves high classification accuracies for both 2-class and 4-class datasets by extracting deep features from MR images and utilizing a novel feature selection algorithm.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Artificial Intelligence
Kursat Demir, Mustafa Ay, Mehmet Cavas, Fatih Demir
Summary: A new deep learning-based approach has been developed in this study to detect and classify surface defects in the steel production process. The proposed methodology involves designing a deep learning model, extracting deep features, selecting features using a new algorithm, and classification using the support vector machine algorithm.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Biomedical
Fatih Demir, Yaman Akbulut, Burak Tasci, Kursat Demir
Summary: A new deep learning model is proposed for brain tumor classification using 3D MRI data. By designing attention, convolutional, and LSTM structures in the same learning architecture, the representation power of the features is increased. The model directly uses 3D MR images and achieves accuracies of 98.90% and 99.29% on the BRATS 2015 and 2018 datasets, respectively.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Fatih Demir, Abdulkadir Sengur, Ali Ari, Kamran Siddique, Mohammed Alswaitti
Summary: A novel approach for Parkinson's disease diagnosis based on speech disorders was developed, which achieved high classification accuracy without feature selection. The deep LSTM network performed both feature extraction and classification processes in an end-to-end manner, resulting in better performance compared to existing methods.
Article
Engineering, Electrical & Electronic
Fatih Demir, Nebras Sobahi, Siuly Siuly, Abdulkadir Sengur
Summary: This study aims to develop an efficient deep feature extraction based method for automatically classifying people's emotional status. The experimental results demonstrate that AlexNet features combined with Alpha rhythm perform better in valence discrimination, while MobilNetv2 features yield the highest accuracy score in arousal discrimination.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Information Systems
Fatih Demir, Aras Masood Ismael, Abdulkadir Sengur
Article
Computer Science, Information Systems
Fatih Demir, Daban Abdulsalam Abdullah, Abdulkadir Sengur
Article
Medical Informatics
Fatih Demir, Abdulkadir Sengur, Varun Bajaj, Kemal Polat
HEALTH INFORMATION SCIENCE AND SYSTEMS
(2019)