Article
Computer Science, Artificial Intelligence
Wenbin Pei, Bing Xue, Lin Shang, Mengjie Zhang
Summary: Due to the imbalance in class distribution, classifiers often exhibit biased performance towards the majority class and perform poorly on the minority class. This study proposes a new cost-sensitive genetic programming method that utilizes rough set theory to detect overlapping areas and improve classification performance on unbalanced high-dimensional data.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Arvind Kumar
Summary: The study examines the impact of imbalanced data on classifier performance and proposes a new fitness function for genetic programming to address imbalanced data classification. Experimental results show that the GP method with the new fitness function outperforms traditional methods and KNN in classifying imbalanced problems.
JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Wenbin Pei, Bing Xue, Lin Shang, Mengjie Zhang
Summary: This paper introduces a novel GP method for developing cost-sensitive classifiers by automatically learning cost matrices instead of requiring them from domain experts, demonstrating superior performance compared to existing GP methods in most cases.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Wenlong Fu, Bing Xue, Xiaoying Gao, Mengjie Zhang
Summary: This paper introduces a transductive transfer learning method for document classification using two different text feature representations to share knowledge and improve performance. It shows that programs learned from TF and doc2vec can be alternatively used to improve each other. Furthermore, it addresses the unbalanced dataset problem by considering the unbalanced distributions on categories.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Divya Acharya, Nandana Varshney, Anindiya Vedant, Yashraj Saxena, Pradeep Tomar, Shivani Goel, Arpit Bhardwaj
Summary: This study focuses on automatic human emotion recognition using EEG signals and analyzes the impact of different genres, age, and gender on human emotions. The results show that the enhanced D-score Genetic Programming performs the best in classifying emotions, and the method's generalizability and reliability are validated through publicly available EEG datasets. The research also reveals that participants in the amusement genre exhibit positive emotions, with brain signals of the 26-35 age group showing the highest emotional identification, and females are more emotionally active compared to males. These findings confirm the potential utility of the method for emotion recognition.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Wenbin Pei, Bing Xue, Lin Shang, Mengjie Zhang
Summary: This article proposes a new fitness function and a multi-criteria selection method to address the bias issue of genetic programming in high-dimensional unbalanced classification. Experimental results show that the proposed method achieves better classification performance than other methods.
EVOLUTIONARY COMPUTATION
(2022)
Article
Computer Science, Information Systems
Ying Bi, Bing Xue, Mengjie Zhang
Summary: This paper proposes a new approach for face image classification based on multi-objective genetic programming. It automatically evolves image descriptors that extract effective features by optimizing both accuracy and distance measure, aiming to enhance generalization. Experimental results on multiple datasets demonstrate that this method significantly outperforms other competitive methods.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Interdisciplinary Applications
Arvind Kumar, Nishant Sinha, Arpit Bhardwaj, Shivani Goel
Summary: Chronic kidney disease is a serious health concern affecting millions of Americans, and early diagnosis through machine learning can help prevent loss of life. This study proposes a new fitness function for dealing with imbalanced data sets in genetic programming, which outperforms other classification techniques in terms of accuracy and AUC values.
COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING
(2022)
Article
Chemistry, Multidisciplinary
Tatiana Avdeenko, Konstantin Serdyukov
Summary: This paper investigates an approach to intelligent support of software white-box testing process using an evolutionary paradigm, solving the problem of automated generation of optimal test data set. By formulating a fitness function with two terms and implementing genetic algorithms, it is possible to achieve maximum statement coverage and population diversity in one launch of the GA. The optimal relation between the two terms of fitness function was obtained for two different programs under testing.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Tingyang Wei, Wei-Li Liu, Jinghui Zhong, Yue-Jiao Gong
Summary: This article proposes an ensemble-based genetic programming classification framework, SE-GEP, which addresses the multiclass classification problem on high dimension and low sample size (HDLSS) data. SE-GEP achieves better classification accuracy compared to other genetic programming methods, and is competitive with other representative machine learning methods for multiclass classification in HDLSS data.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Deepak Kumar Panda, Saptarshi Das, Stuart Townley
Summary: This paper applies machine learning methods to obtain the locational information of energy consumers based on their historical energy consumption patterns. The author tackles the issue of unbalanced classification problem for the dataset and uses Monte Carlo based under-sampling and genetic programming optimizer to optimize and compare the classification algorithms. The classification performance metrics are evaluated and the energy policy implications for urban and rural consumers are discussed.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Review
Computer Science, Information Systems
Le Wang, Meng Han, Xiaojuan Li, Ni Zhang, Haodong Cheng
Summary: This paper explores the classification of unbalanced data sets, analyzing various methods from data sampling, algorithm, feature, cost-sensitive function, and deep learning perspectives, comparing the advantages and disadvantages of different techniques, and outlining future research directions.
Article
Computer Science, Artificial Intelligence
Zhe Ding, Yu-Chu Tian, You-Gan Wang, Weizhe Zhang, Zu-Guo Yu
Summary: Energy efficiency is critical in data center management and operation, especially in cloud computing. This paper proposes a progressive-fidelity approach for improving the computational efficiency of genetic algorithms used for virtual machine placement in data centers. The approach starts with a low-fidelity genetic algorithm and switches to medium and high-fidelity algorithms for solution refinement, with the energy consumption of data centers as the fitness function. Experimental results show that the progressive-fidelity approach is 50% faster for large-scale data centers while maintaining similar solution quality in terms of energy consumption.
APPLIED SOFT COMPUTING
(2023)
Article
Mechanics
Yiding Liu, Zewen Gu, Darren J. Hughes, Jianqiao Ye, Xiaonan Hou
Summary: This study proposed an innovative method to understand the effect of physical attributes on the fracture modes of adhesively bonded joints by combining Finite Element Analysis, Latin Hypercube Sampling, and Genetic Programming. A dataset of 150 samples was generated and the mixed mode ratios were calculated using Strain Energy Release Rate outputs embedded in Linear Elastic Fracture Mechanics, validated by experimental tests. A GP model was developed and trained to evaluate the early-state failure modes of the joints.
COMPOSITE STRUCTURES
(2021)
Article
Computer Science, Artificial Intelligence
Ying Bi, Bing Xue, Mengjie Zhang
Summary: This article proposes a GP-based approach with a dual-tree representation and a new fitness function to automatically learn image features for FSIC. The results show that the proposed approach achieves significantly better performance than a large number of state-of-the-art methods on various types of FSIC tasks.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)