Article
Biochemistry & Molecular Biology
Wei Wei, Yixue Li, Tao Huang
Summary: This study used RNA-seq data and gene chip data to identify potential biomarkers for colorectal cancer. The SMOTE method was used to address class imbalance, and four feature selection algorithms were used to select genes. Four machine learning algorithms were employed for optimal gene selection and model construction. Interpretable machine learning was used to uncover relationships among the selected genes, and survival analysis revealed significant correlations with prognosis. The study also investigated immune cell proportions and gene mutation rates for the selected biomarkers. The identified biomarkers have implications for personalized therapies and improved clinical outcomes.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Yuefan Xu, Sen Zhang, Wendong Xiao
Summary: The variability in ECG patterns among patients and the class imbalance problem pose challenges in ECG recognition. To address these issues, a novel algorithm called ICC-WKELM is proposed for heartbeat multiclass classification. A compact and discriminative feature set is constructed, and a kernel extreme learning machine (KELM) is used for heartbeat classification. The class imbalance is measured using intra-class coherence (ICC), and a weight assignment strategy is designed for imbalanced arrhythmia classes. The proposed approach achieves high F1 scores and overall accuracy on the MIT-BIH arrhythmia dataset.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Pradip Dhal, Chandrashekhar Azad
Summary: Feature selection is crucial in machine learning, but faces challenges in real-world applications. This paper investigates its framework, models, and methods, classifying and discussing algorithms in different data types and applications.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Mariana Daniel, Rui Guerra, Antonio Brazio, Daniela Rodrigues, Ana Margarida Cavaco, Maria Dulce Antunes, Jose Valente de Oliveira
Summary: This study explores the use of feature engineering for preprocessing in fruit classification, as well as the division and selection of wavelength domain spectra. These methods can improve classification accuracy and reduce over-training. Experimental results show that the proposed method outperforms traditional approaches in accuracy and can identify features with physical chemistry significance.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Mathematics
Sang-Ha Sung, Sangjin Kim, Byung-Kwon Park, Do-Young Kang, Sunhae Sul, Jaehyun Jeong, Sung-Phil Kim
Summary: Research shows that feature selection using EEG data in BCI technology can effectively predict whether individuals correctly detect facial expression changes, with specific EEG features largely influencing the detection of expression changes. Various feature selection methods and machine learning techniques were used to achieve high classification accuracy.
Article
Computer Science, Artificial Intelligence
Qing Wu, Yan-Lin Fu, Dong-Shun Cui, En Wang
Summary: This paper proposes a C-loss-based doubly regularized extreme learning machine to address the overfitting and dimensionality reduction problems in extreme learning machines. The proposed method simultaneously completes feature selection and training processes and achieves better regression results and faster training speed in multiple experiments.
COGNITIVE COMPUTATION
(2023)
Article
Computer Science, Information Systems
Harun Olcay Sonkurt, Ali Ercan Altinoz, Emre Cimen, Ferdi Kosger, Gurkan Ozturk
Summary: This study achieved high accuracy in differentiating bipolar disorder patients from healthy controls by utilizing a broader neurocognitive evaluation and a novel machine-learning algorithm.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2021)
Article
Computer Science, Theory & Methods
Muhammad Attique Khan, Habiba Arshad, Wazir Zada Khan, Majed Alhaisoni, Usman Tariq, Hany S. Hussein, Hammam Alshazly, Lobna Osman, Ahmed Elashry
Summary: The paper proposes a framework for human gait recognition based on deep learning and Bayesian optimization. The framework includes extracting motion regions and training a deep model using optical flow, as well as enhancing video frames. Bayesian optimization is used to select hyperparameters, resulting in motion frames and original frames models. Features from both models are fused using Sq-Parallel Fusion, and an ELM classifier is used for classification. Experimental results on CASIA B and CASIA C datasets achieve average accuracies of 92.04% and 94.97% respectively, outperforming other deep learning networks in terms of accuracy.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Theory & Methods
Lu Zhou, Ye Zhu, Tianrui Zong, Yong Xiang
Summary: The paper proposes a DDoS attack flow classification system called SAFE, which accurately and quickly identifies attack flows in the network layer. The proposed method achieves better classification performance in terms of accuracy and efficiency compared to existing methods.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Neurosciences
Fulai Peng, Cai Chen, Danyang Lv, Ningling Zhang, Xingwei Wang, Xikun Zhang, Zhiyong Wang
Summary: This paper proposes a method for gesture recognition based on surface electromyography (sEMG) signals. By combining feature selection and ensemble extreme learning machine (EELM), the recognition performance is significantly improved. The experimental results demonstrate that the proposed method outperforms other methods in terms of accuracy.
FRONTIERS IN HUMAN NEUROSCIENCE
(2022)
Article
Toxicology
Jian Jiang, Jonas van Ertvelde, Goekhan Ertaylan, Ralf Peeters, Danyel Jennen, Theo M. de Kok, Mathieu Vinken
Summary: Drug-induced intrahepatic cholestasis (DIC) is a challenging hepatic toxicity that is difficult to predict in early drug development stages. In vitro toxicogenomics assays using human liver cells have proved to be a practical approach for predicting DIC. This study applied machine learning algorithms to identify transcriptomic signatures of DIC and developed a prediction model with high accuracy and sensitivity. The identified genes provide insights into the mechanisms of DIC and enhance the predictive accuracy of DIC, contributing to the advancement of hazard identification methodologies.
ARCHIVES OF TOXICOLOGY
(2023)
Article
Ecology
Dimitrios Effrosynidis, Avi Arampatzis
Summary: The study found that the wrapper methods SHAP and Permutation Importance are the most effective, while filter methods perform poorly and embedded methods are intermediate. LightGBM performed better among the two machine learning algorithms used. The ensemble method Reciprocal Ranking outperformed all other methods and showed high stability.
ECOLOGICAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Usman Ahmed, Jerry Chun-Wei Lin, Gautam Srivastava, Muhammad Aleem
Summary: This paper presents an approach to map OpenCL applications to heterogeneous multi-core architecture by using a machine learning-based classifier. By selecting features and comparing performance metrics, it achieved efficient classification results. The optimized method was compared with traditional algorithms and applied on AMD and Polybench benchmarks.
Article
Computer Science, Artificial Intelligence
Victor Hugo da Silva Muniz, Joao Baptista de Oliveira e Souza Filho
Summary: This paper discusses the importance of music genre in music recommendations and presents a method to improve system performance through the generation of new handcrafted features and feature selection.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Audi Albtoush, Manuel Fernandez-Delgado, Eva Cernadas, Senen Barro
Summary: Quick Extreme Learning Machine (QELM) is a method that can handle large classification datasets by avoiding tuning and replacing training patterns in the activation matrix with a reduced set of prototypes, thus avoiding the storage and computation of large matrices. It can be executed on general purpose computers within reasonable times and achieves performances similar to extreme learning machine (ELM).
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Jayesh George Melekoodappattu, Perumal Sankar Subbian
JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING
(2020)
Article
Engineering, Electrical & Electronic
Jayesh George Melekoodappattu, Perumal Sankar Subbian, M. P. Flower Queen
Summary: Mammography is an essential technique for diagnosing breast malignancy in women, with mass lesions and microcalcifications being the most common features associated with breast tumors. The Glowworm Swarm Optimization (GSO) algorithm is effective in optimizing feature sets obtained from multiscale feature extraction procedures. The system developed using GSO-ELM-FOA can accurately detect calcifications and tumors with a high level of precision.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2021)
Article
Engineering, Electrical & Electronic
Jayesh George Melekoodappattu, Anoop Balakrishnan Kadan, V Anoop
Summary: This study establishes a computer-aided diagnostic system to interpret breast mammograms, utilizing different wavelet families for feature extraction, employing ANN, SVM, and ELM classifiers for accurate classification, and enhancing middle layer performance through the ELM-GOA algorithm. The results show that the ELM-GOA system can accurately identify breast tumors with a high level of precision.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2021)
Article
Engineering, Electrical & Electronic
Jayesh George Melekoodappattu, Anto Sahaya Dhas, Binil K. Kumar, K. S. Adarsh
Summary: Breast cancer is detected using medical image processing techniques and deep learning methods for automatic detection of malignancy. The ensemble approach shows excellent performance in improving classification efficiency and measurement metrics.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Jayesh George Melekoodappattu, Anto Sahaya Dhas, Binil Kumar Kandathil, K. S. Adarsh
Summary: Customized deep neural networks and image texture attribute extraction are used in this study to autonomously diagnose cancer based on digital mammography images. The findings show that the combination method improves the accuracy and specificity of classification.
JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING
(2022)
Article
Biochemistry & Molecular Biology
Jayesh George Melekoodappattu, Chaithanya Kandambeth Puthiyapurayil, Anoop Vylala, Anto Sahaya Dhas
Summary: This manuscript presents an advanced approach that combines multimodal feature fusion and dual-path network. By leveraging pretrained models and a custom convolutional neural network, salient features are effectively extracted from the data using nonlinear mapping and expansive perception. The resulting two-stage ensemble hybrid CNN model achieves a high accuracy of 99.63% in brain tumor classification.
CELL BIOCHEMISTRY AND FUNCTION
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Kshema, Jayesh M. George, D. Anto Sahaya Dhas
2017 CONFERENCE ON EMERGING DEVICES AND SMART SYSTEMS (ICEDSS)
(2017)