Article
Computer Science, Information Systems
Yushan Wu, Rui Wu, Jiafeng Liu, Xianglong Tang
Summary: Cluster ensemble (CE) integrates multiple clustering solutions to improve unsupervised clustering, but existing methods lack the ability to adaptively adjust weights for different datasets. To address this, we propose Meta-learning-based Weighted Cluster Ensemble (MetaWCE), which automatically learns the weights-data relation to set adaptive CE weights. Experimental results demonstrate that MetaWCE significantly improves ensemble performance compared to baseline methods.
INFORMATION SCIENCES
(2023)
Article
Biochemical Research Methods
Lun Hu, Zhenfeng Li, Zehai Tang, Cheng Zhao, Xi Zhou, Pengwei Hu
Summary: In this study, an ensemble learning algorithm called EM-HIV is proposed for predicting HIV-1 PR cleavage sites. By training a set of weak learners with the asymmetric bagging strategy, EM-HIV can alleviate the impact of data imbalance and noisy data. The algorithm utilizes multiple features from substrate sequences and outperforms state-of-the-art prediction algorithms.
BMC BIOINFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Bin Yu, Xue Wang, Yaqun Zhang, Hongli Gao, Yifei Wang, Yushuang Liu, Xin Gao
Summary: In this study, a deep learning-based framework called RPI-MDLStack is developed for predicting RNA-protein interactions (RPI). By optimizing feature extraction and using a stacking ensemble strategy, RPI-MDLStack achieves high prediction accuracy.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Himanshi Meswal, Deepika Kumar, Aryan Gupta, Sudipta Roy
Summary: This study proposes an effective strategy for accurately classifying skin lesions using a weighted ensemble approach. The proposed classifier achieved high accuracy and performance in the experiments, demonstrating potential for practical application.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Yeou-Ren Shiue, Gui-Rong You, Chao-Ton Su, Hua Chen
Summary: A new ensemble learning approach, ELBAD, based on balanced accuracy and diversity using a two-phase artificial bee colony (ABC) algorithm, is proposed to balance the accuracy and diversity of ensemble learners. Experimental results show that ELBAD significantly outperforms other popular ensemble learning algorithms on multiple datasets.
APPLIED SOFT COMPUTING
(2021)
Article
Agriculture, Multidisciplinary
Kento Koyama, Suxing Lyu
Summary: In previous studies, machine learning models typically used majority vote or average value to predict agricultural product freshness. However, this study considers the subjective nature of freshness evaluation and predicts distributions of spinach leaf freshness by incorporating human uncertainty. The models achieved high performance, with similarity metrics between human freshness evaluation and output distribution indicating realistic predictions.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Optics
Tie Zhang, Jingfu Zheng, Yanbiao Zou
Summary: In order to reduce the deviation in the imaging dimension of workpieces in monocular vision systems and improve the accuracy of dimensional measurement, a BRR-SVR-BPNN weighted voting ensemble learning deviation prediction algorithm is proposed. The algorithm combines the features of probability inference, maximum interval penalty, and nonlinear expression to achieve accurate prediction of deviation. The experimental results demonstrate the effectiveness and robustness of the algorithm, making it easily applicable in industrial settings for improving workpiece dimension measurement accuracy.
OPTICS AND LASER TECHNOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Eunhye Kim, Tsatsral Amarbayasgalan, Hoon Jung
Summary: This study proposes a Multilayer Perceptron-based weighted ensemble method for predicting the accepted parcel volumes during special periods. The experimental study on the dataset provided by Korea Post shows better performance than other compared methods.
APPLIED SCIENCES-BASEL
(2022)
Article
Mathematics
Altyeb Taha
Summary: The continuous development of network technologies has led to the emergence of phishing websites as a major cybersecurity threat. Accurate detection of phishing websites is challenging and ensemble methods are considered state-of-the-art solutions. This paper proposes an intelligent ensemble learning approach based on weighted soft voting, achieving high accuracy in phishing website detection.
Article
Engineering, Environmental
Longxiang Li, Annelise J. Blomberg, Rebecca A. Stern, Choong-Min Kang, Stefania Papatheodorou, Yaguang Wei, Man Liu, Adjani A. Peralta, Carolina L. Z. Vieira, Petros Koutrakis
Summary: A machine learning model was developed to predict monthly radon concentrations for each ZIP Code Tabulation Area in the Greater Boston area based on short-term measurements. A two-stage ensemble-based model showed good prediction accuracy, demonstrating the potential for use in future epidemiological studies.
ENVIRONMENTAL SCIENCE & TECHNOLOGY
(2021)
Article
Genetics & Heredity
Zhong-Hao Ren, Chang-Qing Yu, Li-Ping Li, Zhu-Hong You, Yong-Jian Guan, Yue-Chao Li, Jie Pan
Summary: Non-coding RNAs (ncRNAs) play important roles in biological processes through interactions with RNA binding proteins (RBPs). Computational methods have been developed to predict ncRNA-protein interactions, but some of them have limited applicability. In this study, a computational method called SAWRPI is proposed to predict ncRNA-protein interactions using sequence information. The method achieved high performance in experiments, showing its potential as a reliable tool for predicting ncRNA-protein interactions.
FRONTIERS IN GENETICS
(2022)
Article
Multidisciplinary Sciences
Guochao Wei, Naseer Iqbal, Valentine V. Courouble, Ashwanth C. Francis, Parmit K. Singh, Arpa Hudait, Arun S. Annamalai, Stephanie Bester, Szu-Wei Huang, Nikoloz Shkriabai, Lorenzo Briganti, Reed Haney, Vineet N. KewalRamani, Gregory A. Voth, Alan N. Engelman, Gregory B. Melikyan, Patrick R. Griffin, Francisco Asturias, Mamuka Kvaratskhelia
Summary: This study highlights the importance of prion-like low complexity domains in binding and increasing the avidity when interacting with viral capsid, through structural, biochemical, and virological assays.
NATURE COMMUNICATIONS
(2022)
Article
Multidisciplinary Sciences
Mulagala Sandhya, Mahesh Kumar Morampudi, Rushali Grandhe, Richa Kumari, Chandanreddy Banda, Nagamani Gonthina
Summary: This study proposes a deep learning model for automated detection of the severity of diabetic retinopathy. By using ensembles of pretrained models and data augmentation techniques, the model achieves fair accuracy and shows potential for real-time diagnosis.
Article
Computer Science, Information Systems
Sinam Ajitkumar Singh, Ningthoujam Dinita Devi, Khuraijam Nelson Singh, Khelchandra Thongam, Balakrishna Reddy, Swanirbhar Majumder
Summary: Heart sound signal analysis is a crucial area in healthcare, and this study introduces a transfer learning convolutional neural network (CNN) model-based ensemble learning algorithm to predict imbalanced heart sound signals. By employing spectrogram images and STFT to extract relevant features from Phonocardiogram (PCG) data, the model improves prediction performance. The proposed algorithm outperforms existing methods and opens up potential avenues for future exploration in cardiac diagnostics.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Genetics & Heredity
Zhenfeng Li, Lun Hu, Zehai Tang, Cheng Zhao
Summary: Understanding HIV-1 protease substrate specificity is crucial for HIV infection prevention. A novel positive-unlabeled learning algorithm, PU-HIV, has been proposed for effective prediction of HIV-1 protease cleavage sites, demonstrating superior performance compared to existing prediction models in terms of AUC, PR-AUC, and F-measure. With PU-HIV, previously unknown substrate sites can be identified, offering valuable insights for designing novel HIV-1 protease inhibitors for HIV treatment.
FRONTIERS IN GENETICS
(2021)