Semi-supervised adaptive kernel concept factorization

Wenhui WU, Junhui HOU, Shiqi WANG, Sam KWONG, Yu ZHOU

Research output: Journal PublicationsJournal Article (refereed)peer-review

2 Citations (Scopus)


Kernelized concept factorization (KCF) has shown its advantage on handling data with nonlinear structures; however, the kernels involved in the existing KCF-based methods are empirically predefined, which may compromise the performance. In this paper, we propose semi-supervised adaptive kernel concept factorization (SAKCF), which integrates the data representation and kernel learning into a unified model to make the two learning processes adapt to each other. SAKCF extends traditional KCF in a semi-supervised manner, which encourages the high-dimensional representation to be consistent with both the limited supervisory and local geometric information. Besides, an alternating iterative algorithm is proposed to solve the resulting constrained optimization problem. Experimental results on six real-world data sets verify the effectiveness and advantages of our SAKCF over state-of-the-art methods when applied on the clustering task.
Original languageEnglish
Article number109114
JournalPattern Recognition
Early online date23 Oct 2022
Publication statusPublished - Feb 2023
Externally publishedYes

Bibliographical note

Funding Information:
This work was supported in part by the National Natural Science Foundation of China (Grants 62006158 and 62176160 ), in part by the Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA ), in part by the Hong Kong GRF-RGC General Research Fund under Grant 11209819 (CityU 9042816) and Grant 11203820 (9042598), and in part by the Natural Science Foundation of Shenzhen (University Stability Support Program nos. 20200810150732001).

Publisher Copyright:
© 2022 Elsevier Ltd


  • Clustering
  • Concept factorization
  • Kernel method
  • Nonnegative matrix factorization
  • Semi-supervised learning


Dive into the research topics of 'Semi-supervised adaptive kernel concept factorization'. Together they form a unique fingerprint.

Cite this