Article
Engineering, Multidisciplinary
Yi Liu, Changyun Miao, Xianguo Li, Jianhua Ji, Dejun Meng
Summary: The proposed method combines sound and thermal infrared image features to detect different types of faults in idlers, showing high reliability and easy implementation.
Article
Computer Science, Information Systems
Faizan E. Mustafa, Abdul Qayyum Khan, Abdus Samee, Ijaz Ahmed, Muhammad Abid, Mohammad M. Alqahtani, Muhammad Khalid
Summary: Industrial processes are complex and nonlinear, making it difficult to identify and isolate defects. This study investigates the use of Principal Component Analysis (PCA), Fisher Discriminant Analysis (FDA), Kernel Fisher Discriminant Analysis (KFDA), and Sequential quadratic programming (SQP) for fault detection and isolation in industrial processes. The findings suggest that FDA, KFDA, and SQP are effective in identifying and isolating faults, while PCA has limitations. Data-driven design approaches offer higher reliability and efficiency compared to PCA-based methods in complicated industrial processes.
Article
Environmental Sciences
Xiaojing Wang, Faming Huang, Xuanmei Fan, Himan Shahabi, Ataollah Shirzadi, Huiyuan Bian, Xiongde Ma, Xinxiang Lei, Wei Chen
Summary: This study applies three advanced landslide susceptibility models to evaluate landslide susceptibility in Muchuan County, China. By analyzing the local geo-environmental characteristics and a landslide inventory map, the best model is determined and valuable information for slope stability is provided for local governments and organizations.
ENVIRONMENTAL EARTH SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Chun-Na Li, Yi-Fan Qi, Da Zhao, Tingting Guo, Lan Bai
Summary: 2DLDA is an extension of LDA that can handle matrix input samples directly. However, it is sensitive to noise and outliers. In this paper, a square-free F-norm 2DLDA is proposed to improve its robustness. By eliminating the squared operation, the proposed method weakens the influence of outliers and noise while preserving the geometric structure of data.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Information Systems
Sangamesh Hosgurmath, Viswanatha Vanjre Mallappa, Nagaraj B. Patil, Vishwanath Petli
Summary: The research introduces a new Dual Linear Collaborative Discriminant Regression Classification (DLCDRC) algorithm, which enhances face recognition performance through deep loss function and combined distance metric steps, achieving high classification accuracy on YALE B, ORL, and extended YALE B face datasets.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Muhammad Aminu, Noor Atinah Ahmad
Summary: By incorporating a locality preserving feature, LPPLSDA enhances the performance of partial least squares discriminant analysis, especially in face recognition tasks. Experimental results consistently show that LPPLSDA outperforms the conventional PLS-DA method on various benchmarked face databases.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Si-Tao Ling, Yi-Ding Li, Bing Yang, Zhi-Gang Jia
Summary: This article presents a new joint diagonalization algorithm for a pair of Hermitian quaternion matrices and proposes a two-dimensional quaternion linear discriminant analysis (2D-QLDA) method based on this algorithm for color face recognition and image reconstruction. The method outperforms other methods in color face recognition and image reconstruction.
Article
Computer Science, Artificial Intelligence
M. MaryHelta Daisy, P. Kannan
Summary: This paper proposes a new approach for face recognition by enhancing face representation effectiveness through rotation algorithm and Gabor features, along with adding Cell-based Fishers linear discriminant to reduce dimensionality. The study investigates the impact of different Gabor filter parameters on face recognition accuracy, concluding that Rotated Local Gabor Features provide the best face recognition accuracy.
JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING
(2021)
Article
Computer Science, Information Systems
I. Michael Revina, W. R. Sam Emmanuel
Summary: Facial expression recognition is a powerful tool for social communication, involving preprocessing, feature extraction, and classification stages, with performance of different FER techniques compared based on the number of expressions recognized and algorithm complexity.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2021)
Article
Computer Science, Hardware & Architecture
Yanli Ren, Zhuhuan Song, Shifeng Sun, Joseph K. Liu, Guorui Feng
Summary: This paper proposes a protocol for outsourcing LDA-based face recognition tasks to an untrusted cloud, allowing clients to complete matrix inversion, matrix multiplication, and eigenvalue decomposition operations. The protocol ensures the privacy of client data and allows the client to verify the correctness of the outsourcing results. Additionally, it reduces the computational complexity for the client, enabling efficient execution of the LDA algorithm.
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Chin-Chun Chang
Summary: Fisher's linear discriminant analysis (LDA) is a supervised dimensionality reduction method that may be ineffective against complicated class distributions. This paper demonstrates that deep neural networks with rectified linear units can reveal classification information in a subspace where LDA cannot find any. Combining LDA with space-folding operations and fine-tuning can enhance the classification performance. Experimental results on artificial and open data sets validate the feasibility of the proposed approach.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Haifa Nakouri
Summary: Dimensionality reduction is crucial in face recognition, with SDA being a common method but facing issues of high time and space cost. Recent research suggests that 2D matrix-based methods outperform traditional 1D vector-based ones in performance.
PATTERN ANALYSIS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Huan Wan, Hui Wang, Bryan W. Scotney, Jun Liu, Xin Wei
Summary: The paper proposes a novel variant of subclass-based linear discriminant analysis (LDA) called Global Subclass Discriminant Analysis (GSDA) to address the limitation of traditional LDA in utilizing the locality information in data. GSDA selects subclasses from global clusters that may cross class boundaries, effectively utilizing both within-class and between-class information. Experimental results show that GSDA outperforms state-of-the-art LDA algorithms in terms of accuracy and run times.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Information Systems
Cen Chen, Yun Yang, Xuerong Ye, Guofu Zhai
Summary: This article develops a new type of dictionary URV-LDA dictionary by combining the unit residual signal vector and the linear discriminant analysis for feature transformation, which can better solve the soft faults issues with significant increases on diagnostic accuracy in electromechanical systems.
Article
Computer Science, Artificial Intelligence
Chao Zhang, Huaxiong Li, Chunlin Chen, Xianzhong Zhou
Summary: This paper proposes a novel unsupervised dimensionality reduction method NRDP, which focuses on the local and nonlocal structure of data by simultaneously maximizing nonlocal scatter and minimizing local scatter to learn discriminant projection. NRDP utilizes a nonnegative representation model and l(1)-norm as metric for robustness against noises, and solves the optimization model through an iterative algorithm, significantly improving the representation power and discrimination of features.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2021)
Article
Engineering, Industrial
Sahand Hajifar, Hongyue Sun, Fadel M. Megahed, L. Allison Jones-Farmer, Ehsan Rashedi, Lora A. Cavuoto
Summary: This study examined the use of time series methods to forecast physical fatigue using subjective ratings of perceived exertion and gait data from wearable sensors. The results indicated that the VECM model, incorporating historical RPE and wearable sensor data, outperformed other models in forecasting three or more time periods ahead.
APPLIED ERGONOMICS
(2021)
Article
Transportation Science & Technology
Miao Cai, Mohammad Ali Alamdar Yazdi, Amir Mehdizadeh, Qiong Hu, Alexander Vinel, Karen Davis, Hong Xian, Fadel M. Megahed, Steven E. Rigdon
Summary: This study examines the association between safety-critical events (SCEs) and crashes, injuries, and fatalities among commercial truck drivers, finding a positive correlation between SCEs and accidents and injuries. The results are consistent across different business units and driver types.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Engineering, Electrical & Electronic
Saeb Ragani Lamooki, Jiyeon Kang, Lora A. Cavuoto, Fadel M. Megahed, L. Allison Jones-Farmer
Summary: Gait analysis is traditionally done visually by trained professionals, but there is a growing trend to use features extracted from sensing data as inputs to machine learning methods. This paper introduces a personalized statistical framework for detecting and interpreting individual gait changes.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Industrial
Miao Cai, Amir Mehdizadeh, Qiong Hu, Mohammad Ali Alamdar Yazdi, Alexander Vinel, Karen C. Davis, Hong Xian, Fadel M. Megahed, Steven E. Rigdon
Summary: This study examines the impact of driving shifts and rest breaks on safety critical events in the trucking industry. Findings show that hard brake intensity decreases throughout a shift, rest breaks reduce activation of collision mitigation systems, and there is significant variability among drivers. The study also highlights the importance of proper rest-break scheduling for trucking safety.
JOURNAL OF QUALITY TECHNOLOGY
(2022)
Article
Ergonomics
Amir Mehdizadeh, Mohammad Ali Alamdar Yazdi, Miao Cai, Qiong Hu, Alexander Vinel, Steven E. Rigdon, Karen Davis, Fadel M. Megahed
Summary: This study utilized data from over 20 million miles of driving to address the prediction of driving risks, successfully predicting safety critical events 30 minutes in advance with a relative stability of predictive models and applicability to new drivers.
ACCIDENT ANALYSIS AND PREVENTION
(2021)
Article
Engineering, Manufacturing
David Romero, Thorsten Wuest, Makenzie Keepers, Lora A. Cavuoto, Fadel M. Megahed
Summary: This paper discusses the potential of smart wearable and collaborative technologies in digital manufacturing to create safer and more efficient shop floor environments, emphasizing the need for digital transformation and providing recommendations for manufacturing enterprises to ensure sustainable operation during the coronavirus pandemic and in the Industry 4.0 era.
SMART AND SUSTAINABLE MANUFACTURING SYSTEMS
(2021)
Editorial Material
Biochemical Research Methods
Fadel M. Megahed, Ying-Ju Chen, Aly Megahed, Yuya Ong, Naomi Altman, Martin Krzywinski
Article
Multidisciplinary Sciences
Fadel M. Megahed, L. Allison Jones-Farmer, Longwen Zhao, Steven E. Rigdon
Summary: The study identified three distinct patterns of COVID-19 cases across counties in the U.S., with significant associations with demographic, socioeconomic, and political variables. The outbreak patterns are related to geographic location within the U.S. and important factors such as population density.
Article
Chemistry, Analytical
Sahand Hajifar, Saeb Ragani Lamooki, Lora A. Cavuoto, Fadel M. Megahed, Hongyue Sun
Summary: This study investigates the impact of four heterogeneity sources on classification performance and conducts experiments to simulate tasks of electrical line workers. The support vector machine equipped with domain adaptation outperforms the baseline in certain cases, but does not perform better in the cross-scenario case.
Article
Computer Science, Interdisciplinary Applications
Hamidreza Ahady Dolatsara, Ying-Ju Chen, Robert D. Leonard, Fadel M. Megahed, L. Allison Jones-Farmer
Summary: This article evaluates the impact of data preparation and model selection on the predictive accuracy of models applied to a heart transplantation database. It highlights the interactions between early and later decisions and emphasizes the need for improved rigor in applied predictive research.
Article
Engineering, Industrial
Saeb Ragani Lamooki, Sahand Hajifar, Jiyeon Kang, Hongyue Sun, Fadel M. Megahed, Lora A. Cavuoto
Summary: This study investigates the effectiveness of an end-to-end framework that utilizes data from a single wearable sensor for ergonomic risk assessment. By identifying tasks and estimating task intensity, the framework eliminates the need for direct observation and achieves good accuracy rates and estimation errors.
APPLIED ERGONOMICS
(2022)
Article
Public, Environmental & Occupational Health
Fadel M. Megahed, L. Allison Jones-Farmer, Yinjiao Ma, Steven E. Rigdon
Summary: This study aimed to determine the number of distinct clusters of COVID-19 deaths in 3108 contiguous counties in the United States, their geographical distribution, and the factors influencing cluster membership. The findings suggest that county-level patterns of COVID-19 deaths vary and can be partly explained by social and political predictors.
JMIR PUBLIC HEALTH AND SURVEILLANCE
(2022)
Article
Multidisciplinary Sciences
Saeb Ragani Lamooki, Sahand Hajifar, Jacqueline Hannan, Hongyue Sun, Fadel Megahed, Lora Cavuoto
Summary: This study presents a new method for identifying tasks performed by electrical line workers using wrist-worn accelerometers. The data from 37 participants in a lab environment were analyzed and achieved high accuracy in task classification. This research is important for future risk mitigation of electrical line workers using wearables.
Article
Engineering, Industrial
Fadel M. Megahed, Ying-Ju Chen, Joshua A. Ferris, Sven Knoth, L. Allison Jones-Farmer
Summary: Generative AI models like OpenAI's ChatGPT have the potential to revolutionize Statistical Process Control (SPC) practice, learning, and research. However, their early stage of development makes them susceptible to misuse and misunderstandings. In this paper, we examine the capabilities of ChatGPT in providing code, explaining concepts, and generating knowledge related to SPC. While it performs well in structured tasks, it struggles with more nuanced ones. The use of generative AI models in SPC requires proper validation and should be complemented with other methods for accurate results.
QUALITY ENGINEERING
(2023)
Article
Computer Science, Information Systems
Jane F. Moore, Arthur Carvalho, Gerard A. Davis, Yousif Abulhassan, Fadel M. Megahed
Summary: This paper discusses the importance of implementing social distancing controls in public transit, introduces two types of social distancing models to address the limitations of existing models, and demonstrates the effectiveness and feasibility of these models through illustrative scenarios.