ジャーナル論文 Modularity Regularized Locality Preserving Projection
Ibuki Masuda (author) (この著者で検索)
;
Yasunori Futamura (author) (この著者で検索)
;
Jianbo Lin (author) (この著者で検索)
ORCID https://orcid.org/0000-0003-0769-9857
National Institute for Materials Science
ORCID ;
Anh Khoa Augustin Lu (author) (この著者で検索)
ORCID SAMURAI ;
Ryo Tamura (author) (この著者で検索)
ORCID SAMURAI ;
Tsuyoshi Miyazaki (author) (この著者で検索)
ORCID SAMURAI ;
Tetsuya Sakurai (author) (この著者で検索)
コレクション

引用
Ibuki Masuda, Yasunori Futamura, Jianbo Lin, Anh Khoa Augustin Lu, Ryo Tamura, Tsuyoshi Miyazaki, Tetsuya Sakurai. Modularity Regularized Locality Preserving Projection. Communications in Computer and Information Science. 2025, 2754 (), 528-539. https://doi.org/10.1007/978-981-95-4091-4_36

説明:

(abstract)

Dimensionality reduction is a crucial preprocessing step for clustering. While nonlinear dimensionality reduction methods effectively capture complex data structures, linear dimensionality reduction methods are often preferred in applications requiring interpretability. However, existing linear approaches, while preserving local or global data characteristics, often struggle to effectively preserve cluster structures. In this paper, we propose a novel linear dimensionality reduction method based on modularity maximization, which is widely used in community detection. The proposed method is formulated as an extension of Locality Preserving Projection (LPP), controlled by a single parameter. More precisely, our method preserves local structural information based on the principle of LPP, while incorporating a distinctive regularization term designed to promote inter-cluster separation. We further show that this parameter serves to control the inter-cluster separation. Experimental results demonstrate that our method can robustly obtain a low-dimensional space that preserves cluster structures, particularly for datasets with high-variance clusters, compared to conventional methods.

権利情報:

キーワード: Unsupervised machine learning, Clustering, Local Structure Analysis, Linear Dimensionality Reduction

刊行年月日: 2025-11-09

出版者: Springer Nature Singapore

掲載誌:

  • Communications in Computer and Information Science (ISSN: 18650929) vol. 2754 p. 528-539

研究助成金:

  • JSPS 21H01008 (科研費)
  • JSPS 23K20816 (科研費)
  • JSPS 22K12046 (科研費)
  • JSPS 23H03411 (科研費)

原稿種別: 著者最終稿 (Accepted manuscript)

MDR DOI: https://doi.org/10.48505/nims.6009

公開URL: https://doi.org/10.1007/978-981-95-4091-4_36

関連資料:

その他の識別子:

連絡先:

更新時刻: 2026-11-09 10:30:05 +0900

MDRでの公開時刻: 2026-11-09 10:28:18 +0900

ファイル名 サイズ
ファイル名 Modularity_Regularized_Locality_Preserving_Projection.pdf (サムネイル)
application/pdf
サイズ 2.32MB 詳細