Article
Computer Science, Information Systems
Xiuhua Chen, Chen Gong, Jian Yang
Summary: This paper proposes a novel algorithm CSPU for PU learning, which addresses class imbalance by imposing different misclassification costs on different classes. The algorithm outperforms other comparators in dealing with minority classes.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Shanlin Zhou, Yan Gu, Hualong Yu, Xibei Yang, Shang Gao
Summary: This article introduces a novel strategy called RUE for estimating location information and cost assignment to address the problem of class-imbalance learning. The strategy indirectly explores location information through a random undersampling ensemble, is robust towards data distribution, and accurately estimates the significance of each instance.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Wenbin Chen, Qinghua Zhang, Yongyang Dai
Summary: This study proposes a new sequential multi-class three-way decision model by considering the granular structure of the sequential process. The model defines decision cost, calculates attribute sequence, and the experimental results demonstrate its advantage in decision cost.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2022)
Article
Computer Science, Information Systems
Nan Wang, Yubo Zhang, Xibin Zhao, Yingli Zheng, Hao Fan, Boya Zhou, Yue Gao
Summary: In this paper, a search-based cost-sensitive hypergraph learning method is proposed for anomaly detection. The method effectively addresses the issues of class imbalance and unclear correlation commonly encountered in anomaly detection tasks. Experimental results on industry datasets and other benchmark datasets demonstrate the effectiveness and superiority of the proposed method.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Yilena Perez-Almaguer, Raciel Yera, Ahmad A. Alzahrani, Luis Martinez
Summary: Group recommender systems recommend items consumed socially by groups, using collaborative filtering as the core algorithm. This study explores a taxonomy for content-based group recommendation systems and analyzes three specific models, as well as proposing a hybrid CB-GRS. Experimentation over well-known datasets is conducted to evaluate the proposals and provide a basis for further research in content-based group recommender systems.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Interdisciplinary Applications
Yi Yang, Yuxuan Guo, Xiangyu Chang
Summary: The article introduces a novel angle-based cost-sensitive classification framework for multicategory classification without the sum-to-zero constraint, with loss functions proven to be Fisher consistent. Two cost-sensitive multicategory boosting algorithms derived from the framework show competitive classification performances in numerical experiments against other existing boosting approaches.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2021)
Article
Computer Science, Information Systems
Saban Ozturk, Emin Celik, Tolga Cukur
Summary: The increasing utilization of medical imaging technology has led to the development of large-scale data repositories. Content-based image retrieval (CBIR) provides an automated solution for fast access to similar image samples, and the OCAM method has shown superior performance in medical image querying.
INFORMATION SCIENCES
(2023)
Article
Engineering, Environmental
Chi Peng, Qingfeng Li, Jianhong Fu, Yun Yang, Xiaomin Zhang, Yu Su, Zhaoyang Xu, Chengxu Zhong, Pengcheng Wu
Summary: Kick detection is crucial for drilling operation safety. In this study, a novel intelligent model is proposed for early kick detection, which incorporates feature transformation, cost-sensitive dataset construction, and ensemble learning. The model shows excellent performance in different data dimensions and misclassification costs, and outperforms conventional methods according to ablation experiments and comparisons. The proposed model demonstrates better early kick detection performance than existing methods.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2023)
Article
Education & Educational Research
Joy Jeevamol, V. G. Renumol
Summary: This paper addresses the new user cold-start problem in e-learning content RSs by proposing an ontology-based content recommender system. By incorporating additional learner data in the recommendation process, the proposed model provides more reliable and personalized recommendations.
EDUCATION AND INFORMATION TECHNOLOGIES
(2021)
Article
Computer Science, Artificial Intelligence
Yi Yang, Shuai Huang, Wei Huang, Xiangyu Chang
Summary: This article introduces a method that combines cost-sensitive learning and privacy protection by incorporating weight constants and weight functions, proposing two privacy-preserving algorithms, and demonstrating that this general framework can reduce misclassification costs and meet privacy requirements. Theoretical research shows that the choice of weight constants and weight functions not only affects the algorithm's consistency properties, but also significantly interacts with privacy protection levels.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Theory & Methods
M. Mehdi Afsar, Trafford Crump, Behrouz Far
Summary: Recommender systems have become an integral part of our daily lives, helping us find our favorite items, friends on social networks, and movies to watch. The recommendation problem was traditionally seen as a classification or prediction problem, but it is now widely agreed that formulating it as a sequential decision problem using reinforcement learning can better capture user-system interaction and long-term engagement.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Information Systems
Xinmin Zhang, Saite Fan, Zhihuan Song
Summary: Fault classification is crucial in industrial process monitoring, but the imbalanced distribution of real-life datasets poses challenges. This paper proposes a novel reinforcement learning-based cost-sensitive classifier (RLCC) for imbalanced fault classification. RLCC utilizes a cost-sensitive learning strategy and a newly designed reward, trained through an alternating iterative approach.
SCIENCE CHINA-INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Artittayapron Rojarath, Wararat Songpan
Summary: Ensemble learning combines diverse classification models to improve prediction efficiency, utilizing cost-sensitive learning to enhance accuracy. Experimental results show that the weighted voting ensemble model derived from three models provides the most accurate results in multi-class prediction.
APPLIED INTELLIGENCE
(2021)
Review
Computer Science, Information Systems
Jeevamol Joy, Renumol Vemballiveli Govinda Pillai
Summary: This paper provides insights into recent research trends and technologies in the field of content recommendation in e-learning. Through a literature review, the paper categorizes and analyzes relevant articles on recommendation techniques, data inputs, algorithms, similarity measures, and evaluation metrics.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Naji Ahmad Albatayneh, Khairil Imran Ghauth, Fang-Fang Chua
Summary: This paper presents a preference learning model called Discriminate2Rec, which improves the recommendation accuracy of content-based recommender systems by enhancing the coherence of user profile at the semantic and temporal attribute level. Evaluation on three real-world datasets demonstrates the superior performance of Discriminate2Rec compared to state-of-the-art approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)