Kun Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Kun Xu has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Fatin Izzati Ramli
Universiti Putra Malaysia
1 shared publication - Ge Zhang
Innovation Academy for Microsatellites of CAS, Shanghai 201203, China
1 shared publication - Haibin Zhong
Universiti Putra Malaysia
1 shared publication - Hao Li
School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China
1 shared publication - Mudathir Ahmed Mohamud
Department of Computer Science, Faculty of Computing, SIMAD University, Mogadishu, Somalia
1 shared publication - Pengfei Li
Hangzhou Dianzi University, Baiyang Road #2, Hangzhou, China
1 shared publication
Latest Publications
- 9 September 2026
- Published by Universiti Putra Malaysia in Journal of Intelligent Media Computing
- Vol. 1 (1) , 135
An Efficient Framework for Processing Range Skyline Queries over Incomplete Database
Skyline queries retrieve objects that are not dominated by any other object in the database, playing a key role in preference analysis and multi-criteria decision-making system. As an important variant, range skyline queries retrieve skyline results within a database defined by the range value across one or more dimensions. The presence of incomplete data is inevitable in large-scale, multidimensional databases, introducing new challenges in processing range skyline queries. To address this challenge, an efficient framework named Optimized Incomplete Range Skyline (OIRS) is proposed to optimize range skyline query execution by incorporating a range pruning strategy and a bucket metadata reuse technique. The OIRS framework takes precomputed buckets and their metadata as input and consists of three phases: pruning objects with range query, computing bucket skylines, and deriving final skyline results. First, a range pruning algorithm prunes irrelevant objects based on bucket metadata. Next, a bucket skyline computation algorithm efficiently computes bucket skylines for the remaining objects by leveraging a bucket skyline reuse technique. Finally, the final skyline results are derived from the candidate objects across the bucket skylines. Extensive experiments conducted on both synthetic and real-world NBA datasets demonstrate that OIRS significantly outperforms the naïve scan-and-compute approach across three key metrics: the number of object scans in the range pruning phase, the number of pairwise comparisons in computing bucket skylines, and the number of pairwise comparisons for deriving final incomplete skylines.
- 9 September 2026
- Published by Universiti Putra Malaysia in Journal of Intelligent Media Computing
- Vol. 1 (1) , 135
An Efficient Framework for Processing Range Skyline Queries over Incomplete Database
Skyline queries retrieve objects that are not dominated by any other object in the database, playing a key role in preference analysis and multi-criteria decision-making system. As an important variant, range skyline queries retrieve skyline results within a database defined by the range value across one or more dimensions. The presence of incomplete data is inevitable in large-scale, multidimensional databases, introducing new challenges in processing range skyline queries. To address this challenge, an efficient framework named Optimized Incomplete Range Skyline (OIRS) is proposed to optimize range skyline query execution by incorporating a range pruning strategy and a bucket metadata reuse technique. The OIRS framework takes precomputed buckets and their metadata as input and consists of three phases: pruning objects with range query, computing bucket skylines, and deriving final skyline results. First, a range pruning algorithm prunes irrelevant objects based on bucket metadata. Next, a bucket skyline computation algorithm efficiently computes bucket skylines for the remaining objects by leveraging a bucket skyline reuse technique. Finally, the final skyline results are derived from the candidate objects across the bucket skylines. Extensive experiments conducted on both synthetic and real-world NBA datasets demonstrate that OIRS significantly outperforms the naïve scan-and-compute approach across three key metrics: the number of object scans in the range pruning phase, the number of pairwise comparisons in computing bucket skylines, and the number of pairwise comparisons for deriving final incomplete skylines.
- 21 March 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 34 (09)
Exploring Uncertainty and Representativeness for Deep Active Learning
Albeit the considerable advances that are achieved by convolutional neural networks (CNN) in active learning, simultaneously investing multiple criteria such as uncertainty and representativeness when sampling informative images still remains a challenging issue. In this work, we develop and evaluate an active learning method called Exploring Uncertainty and Representativeness in Deep Active Learning (EURDAL), to iteratively boost the generalization capacity of the CNN backbone for image classification. To be specific, we introduce two adversarial image classifiers appended to the CNN backbone, and the uncertainty is expounded by the prediction discrepancy of the two adversarial classifiers. Besides, the ([ ])-tuplet loss, which was developed in deep metric learning, is imposed on the CNN backbone, to force CNN to extract discriminative features of unlabeled images. After that, the representativeness is defined by the distance proportion between the given example to its own centroid and the given example to all cluster centroids. With the integration of the learned uncertainty and representativeness, the most informative images will be selected while the noisy ones will be suppressed at each iteration of active supervision. Intensive image classification experiments are conducted on three benchmark datasets, and the encouraging results demonstrate the superiority and effectiveness of our proposed EURDAL network compared with some current competitive active learning methods.
- 21 March 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 34 (09)
Exploring Uncertainty and Representativeness for Deep Active Learning
Albeit the considerable advances that are achieved by convolutional neural networks (CNN) in active learning, simultaneously investing multiple criteria such as uncertainty and representativeness when sampling informative images still remains a challenging issue. In this work, we develop and evaluate an active learning method called Exploring Uncertainty and Representativeness in Deep Active Learning (EURDAL), to iteratively boost the generalization capacity of the CNN backbone for image classification. To be specific, we introduce two adversarial image classifiers appended to the CNN backbone, and the uncertainty is expounded by the prediction discrepancy of the two adversarial classifiers. Besides, the ([ ])-tuplet loss, which was developed in deep metric learning, is imposed on the CNN backbone, to force CNN to extract discriminative features of unlabeled images. After that, the representativeness is defined by the distance proportion between the given example to its own centroid and the given example to all cluster centroids. With the integration of the learned uncertainty and representativeness, the most informative images will be selected while the noisy ones will be suppressed at each iteration of active supervision. Intensive image classification experiments are conducted on three benchmark datasets, and the encouraging results demonstrate the superiority and effectiveness of our proposed EURDAL network compared with some current competitive active learning methods.