"Unsupervised random forest for affinity estimation" by Yunai Yi, Diya Sun et al.

Article Title

Unsupervised random forest for affinity estimation

Authors

Yunai Yi, Key Laboratory of Machine Perception (MOE), Department of Machine Intelligence, Peking University, Beijing 100871, China
Diya Sun, Key Laboratory of Machine Perception (MOE), Department of Machine Intelligence, Peking University, Beijing 100871, China
Peixin Li, Key Laboratory of Machine Perception (MOE), Department of Machine Intelligence, Peking University, Beijing 100871, China
Tae-Kyun Kim, Department of Electrical and Electronic Engineering, Imperial College London, London, UK
Tianmin Xu, School of Stomatology, Stomatology Hospital, PekingUniversity, Beijing 100081, China
Yuru Pei, Key Laboratory of Machine Perception (MOE), Department of Machine Intelligence, Peking University, Beijing 100871, China

Keywords

affinity estimation, forest-based metric, unsupervised clustering forest, pseudo-leaf-splitting (PLS)

Abstract

This paper presents an unsupervised cluste-ring random-forest-based metric for affinity estimation in large and high-dimensional data. The criterion usedfor node splitting during forest construction can handle rank-deficiency when measuring cluster compactness. The binary forest-based metric is extended to continuous metrics by exploiting both the common traversal path and the smallest shared parent node.

Recommended Citation

Yi, Yunai; Sun, Diya; Li, Peixin; Kim, Tae-Kyun; Xu, Tianmin; and Pei, Yuru (2022) "Unsupervised random forest for affinity estimation," Computational Visual Media: Vol. 8: Iss. 2, Article 6.
DOI: https://doi.org/10.1007/s41095-021-0241-9
Available at: https://dc.tsinghuajournals.com/computational-visual-media/vol8/iss2/6

Download

Request a Copy

Included in

Computer-Aided Engineering and Design Commons

COinS