Article
Computer Science, Artificial Intelligence
Jihong Wan, Hongmei Chen, Tianrui Li, Wei Huang, Min Li, Chuan Luo
Summary: Information theoretic-guided feature selection approaches aim to select the most informative features. However, previous approaches neglect the complementarity and interaction between features, and fail to fully consider the multi-correlation among features. This study addresses these issues by designing a feature selection algorithm based on class-based relevance, redundancy, complementarity, and interaction, and explores the distinctions and connections among different correlations.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Xi-Ao Ma, Hao Xu, Chunhua Ju
Summary: This paper proposes a class-specific feature selection method based on information theory. A class-specific feature evaluation criterion called CSMDCCMR is developed, and a feature selection algorithm is designed to select a suitable feature subset for each class. Experimental results demonstrate the superiority of the proposed method compared to other methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Heng Liu, Gregory Ditzler
Summary: In adversarial environments, feature selection (FS) in machine learning is vulnerable and has been largely overlooked in research; insecure FS in machine learning pipelines can be the Achilles heel of data pipelines.
INFORMATION SCIENCES
(2021)
Article
Automation & Control Systems
Ping Zhang, Wanfu Gao, Juncheng Hu, Yonghao Li
Summary: Feature selection is crucial in machine learning and data mining, with traditional methods being improved upon by the novel CWJR-FS method, which utilizes conditional-weight joint relevance to design a new feature relevancy term, outperforming other methods in experiments.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Lianxi Wang, Shengyi Jiang, Siyu Jiang
Summary: The study introduces a novel feature selection algorithm that selects relevant and interactive features using a maximum criterion, leading to improved classification accuracy. Experimental results show that the algorithm efficiently selects features and enhances classifiers to achieve better or comparable classification accuracy compared to ten representative competing feature selection algorithms.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Peng Zhou, Peipei Li, Shu Zhao, Xindong Wu
Summary: Existing feature selection methods often overlook the interaction between features, while it has been found that feature interaction plays an important role in influencing the target concept. Therefore, this article introduces a new streaming feature selection method SFS-FI, which considers the interaction between features and evaluates the degree of interaction through a new metric called interaction gain.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Automation & Control Systems
Kshema Shaju, Sherin Babu, Binu Thomas
Summary: This study analyzes the effectiveness of the application of grey theory in feature selection for daily dew point temperature and daily pan-evaporation estimation models. Comparisons and analyses are made between the feature subset identified by grey theory and subsets selected based on different Pearson correlation coefficient slabs. The results show that the models using grey theory-based feature selection demonstrated average or above-average performances.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Physics, Multidisciplinary
Lingbo Gao, Yiqiang Wang, Yonghao Li, Ping Zhang, Liang Hu
Summary: A novel feature selection method, combining three aspects of candidate features, selected features, and label correlations, is proposed in this paper to evaluate feature relevance. By reducing unnecessary redundancy, the method is able to better capture the optimal features and outperform other state-of-the-art multi-label approaches in experiments.
Article
Computer Science, Information Systems
Xi-Ao Ma, Chunhua Ju
Summary: The concepts of feature relevance, redundancy, and complementarity are investigated based on fuzzy information-theoretic measures to handle continuous features. Theoretical definitions and a computationally effective feature evaluation criterion are proposed, and a feature selection algorithm combining the criterion with the sequential forward search strategy is introduced. Experimental results demonstrate the effectiveness of the method.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Benjamin Regler, Matthias Scheffler, Luca M. Ghiringhelli
Summary: In this study, a new measure called total cumulative mutual information (TCMI) is introduced to assess the relevance of random variables with continuous distribution. TCMI is a non-parametric, robust, and deterministic measure that allows for feature selection.
DATA MINING AND KNOWLEDGE DISCOVERY
(2022)
Article
Computer Science, Artificial Intelligence
Keyu Liu, Tianrui Li, Xibei Yang, Hongmei Chen, Jie Wang, Zhixuan Deng
Summary: In this article, a novel semi-supervised feature selection scheme called SemiFREE is proposed, which redefines the feature relevance and redundancy by considering the fuzziness or uncertainty in data labeling. Experimental results demonstrate the superiority of SemiFREE in the presence of partially labeled data.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Biology
Ya-Bian Luo, Yan-Yao Hou, Zhen Wang, Xin-Man Hu, Wei Li, Yan Li, Yong Liu, Tong-Jiang Li, Chun-Zhi Ai
Summary: This study developed machine learning models to predict the metabolic properties of UGT1A1 substrates. The models demonstrated good accuracy and robustness, and were validated with in vitro assays. This strategy is important for optimizing drug metabolism and avoiding drug-drug interactions in clinical practice.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Ghazaala Yasmin, Asit Kumar Das, Janmenjoy Nayak, S. Vimal, Soumi Dutta
Summary: Speech is a delicate medium for identifying the gender of speakers. Deep learning has provided a good research area to explore gender discrimination deficiencies in traditional machine learning techniques. The combination of automatically generated features and human-generated features can enhance gender recognition performance, especially when considering transgender individuals.
Article
Computer Science, Information Systems
Jingliu Lai, Hongmei Chen, Tianrui Li, Xiaoling Yang
Summary: In this study, a novel semi-supervised sparse feature selection framework is proposed, which improves the quality of the similarity matrix through adaptive graph learning and alleviates the negative influence of redundant features through redundancy minimization regularization.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Yuling Fan, Baihua Chen, Weiqin Huang, Jinghua Liu, Wei Weng, Weiyao Lan
Summary: The task of multi-label feature selection (MLFS) is to reduce redundant information and generate the optimal feature subset from the original multi-label data. Existing methods either explore label correlations using pseudo label matrix or consider feature redundancy using information theory technique, but no prior literature unites them in a framework to perform feature selection. Therefore, a novel MLFS method called LFFS is proposed based on label correlations and feature redundancy. Experimental results demonstrate the effectiveness and superiority of the LFFS method among ten competition methods.
KNOWLEDGE-BASED SYSTEMS
(2022)