Article
Computer Science, Artificial Intelligence
Peng Zhou, Xia Wang, Liang Du
Summary: Unsupervised feature selection is an important task in machine learning but suffers from stability and robustness issues due to the absence of labels. This paper proposes a novel bi-level feature selection ensemble method that not only ensembles at the feature level but also learns a consensus clustering result to guide the feature selection, outperforming other state-of-the-art methods.
INFORMATION FUSION
(2023)
Article
Biology
Aiguo Wang, Huancheng Liu, Jing Yang, Guilin Chen
Summary: In this study, an ensemble feature selection framework is proposed to improve the discrimination and stability of features. By using sampling and aggregation strategies, accurate feature selection is achieved in small sample and high dimensionality scenarios, leading to improved diagnostic accuracy and understanding of disease mechanisms.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Gang Wang, Tao Tao, Jingling Ma, Hui Li, Huimin Fu, Yan Chu
Summary: This study proposes a novel ensemble learning method called ALS-RS based on the complementary effect of shallow and deep features for accurate exchange rate forecasting. Experimental results demonstrate the superiority of ALS-RS on four exchange rate datasets and confirm the enhanced forecasting capability of combining multiple features including shallow and deep features.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Biology
Vahid Nosrati, Mohsen Rahmani
Summary: This paper presents a novel framework, named feature-level aggregation-based ensemble based on overlapped feature subspace partitioning (FLAE-OFSP), for microarray data classification. The proposed ensemble generates multiple subsets and applies feature selection algorithms to each subset, and the results are combined into a single ranked list. Evaluation on seven microarray datasets shows significant improvement in runtime and quality results compared to individual methods.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Engineering, Multidisciplinary
Naz Gul, Wali Khan Mashwani, Muhammad Aamir, Saeed Aldahmani, Zardad Khan
Summary: This paper proposes two novel approaches, namely feature weighting and model selection, for building more accurate kNN ensembles. The first approach identifies the nearest observations using a feature weighting scheme based on support vectors. A subset of features is randomly selected for the model construction. After building a sufficient number of base models on bootstrap samples, a subset of the models is selected based on out-of-bag prediction error for the final ensemble. The second approach builds base learners using random subsamples and a random subset of features, incorporating feature weighting. The remaining observations are used to assess the base learners and select a subset of models for the final ensemble. The proposed ensemble methods are compared with other classical methods, including kNN-based models, on 12 benchmark datasets, showing superior performance.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Energy & Fuels
Julian Hoxha, Muhammed Yasin Codur, Enea Mustafaraj, Hassan Kanj, Ali El Masri
Summary: The transportation sector accounts for a significant portion of global oil consumption and energy demand. Accurate prediction of transportation energy consumption is crucial for informed decision-making, sustainable urban transportation, and mitigating carbon emissions. This study proposes a novel methodology involving machine learning techniques to predict transportation energy demand in Turkey and offers insights for future energy investment and policy decisions.
Article
Computer Science, Artificial Intelligence
Ahmad Alsahaf, Nicolai Petkov, Vikram Shenoy, George Azzopardi
Summary: This study introduces a novel feature selection framework based on boosting algorithm for selecting informative feature sets in classification problems. Comparative experiments on benchmark datasets show that the proposed method achieves higher accuracy with fewer features on most datasets, and the selected features exhibit lower redundancy.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Saptarshi Chakraborty, Swagatam Das
Summary: In this paper, a simple and efficient sparse clustering algorithm called LW-k-means is proposed for high-dimensional data. The algorithm incorporates feature weighting to enable feature selection and has a time complexity similar to traditional algorithms. The strong consistency of the LW-k-means procedure is also established. Experimental results on synthetic and real-life datasets demonstrate that LW-k-means performs competitively in terms of clustering accuracy and computational time compared to existing methods for center-based high-dimensional clustering.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Interdisciplinary Applications
Leiming Jin, Wenying Du, Baoqiang Ma, Debin Zeng, Ying Han, Shuyu Li
Summary: This study proposed an FL-GL method by considering correlation at the feature level to improve the traditional multitask learning framework, resulting in better classification performance for patients with amnestic mild cognitive impairment and normal controls.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Computer Science, Artificial Intelligence
Deepak Kumar Rakesh, Raj Anwit, Prasanta K. Jana
Summary: This paper introduces a new frequency-based stability measure called rank stability (RSt), which evaluates feature selection algorithms considering both subsets of features and feature rankings. The proposed measure assesses the variation of feature rankings generated by perturbing the training set. Extensive experiments demonstrate that heterogeneous ensemble techniques outperform traditional feature selection algorithms in terms of the proposed measure and other performance metrics.
Article
Microbiology
Songbo Liu, Chengmin Cui, Huipeng Chen, Tong Liu
Summary: This study introduces an ensemble learning-based method for identifying important features in phage protein, aiming to understand its relationship with host bacteria and develop antimicrobial agents. The selected features are found to have significant biological significance based on the analysis conducted.
FRONTIERS IN MICROBIOLOGY
(2022)
Article
Automation & Control Systems
Ayan Kumar Panja, Syed Fahim Karim, Sarmistha Neogy, Chandreyee Chowdhury
Summary: In this study, a combination of Particle Swarm Optimization and feature-based ensemble model was used to achieve high accuracy in indoor localization while reducing the number of APs needed. The results show that the proposed method can achieve 86%-96% accuracy with a 50%-65% reduction in APs, making it a promising approach for user localization in indoor spaces.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Ting Wu, Yihang Hao, Bo Yang, Lizhi Peng
Summary: Currently, feature selection faces a challenge where no single method can effectively handle various data sets. Ensemble learning is a potential solution, and we propose an ensemble feature selection method based on enhanced co-association matrix (ECM-EFS). We introduce positive-co-association matrix (PCM), negative-co-association matrix (NCM), and relative-co-association matrix (RCM) to discover feature relationships, and use feature kernel for more stable selection. Experimental results show that ECM-EFS provides robust results and reduces computation cost compared to traditional methods.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Sura Emanet, Gozde Karatas Baydogmus, Onder Demir
Summary: This study focuses on developing an advanced Intrusion Detection System (IDS) with high accuracy through feature selection and ensemble learning methods. The researchers reduced the dataset and implemented ensemble learning to optimize IDS performance. The study's significance lies in its contribution to advancing IDS capabilities and improving computer security. The findings validate the effectiveness of ensemble learning in enhancing IDS performance.
PEERJ COMPUTER SCIENCE
(2023)
Article
Construction & Building Technology
Jiaxin Guo, Sining Yun, Yao Meng, Ning He, Dongfu Ye, Zeni Zhao, Lingyun Jia, Liu Yang
Summary: This study proposes four hybrid models (Random-LightGBM, Grid-LightGBM, CMA-ES-LightGBM, and TPE-LightGBM) combined with the LightGBM model for improved prediction accuracy of heating and cooling loads in residential buildings.
BUILDING AND ENVIRONMENT
(2023)
