Multi-View Clustering on Topological Manifold

Authors: Shudong Huang, Ivor Tsang, Zenglin Xu, Jiancheng Lv, Quan-Hui Liu6944-6951

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Substantial experiments on benchmark datasets are conducted to validate the effectiveness of the proposed method, compared to the state-of-the-art algorithms over the clustering performance.
Researcher Affiliation Academia 1 College of Computer Science, Sichuan University, Chengdu 610065, China 2 Centre for Artificial Intelligence, FEIT, University of Technology Sydney, Sydney, NSW 2007, Australia 3 School of Computer Science and Technology, Harbin Institute of Technology Shenzhen, Shenzhen 518055, China
Pseudocode Yes Algorithm 1: Algorithm to solve Eq. (13)
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes Several benchmark multi-view data sets are used in this paper: 3source, Handwritten numerals, Caltech7 and Caltech20.
Dataset Splits No The paper does not provide specific training/validation dataset splits, as is common for unsupervised clustering tasks where the entire dataset is often used for the clustering process.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers needed to replicate the experiment.
Experiment Setup Yes Input: Initial graphs {G(1), G(2), . . . , G(m)} for the m views, cluster number c, parameters α and β. Initialize the weight of each view µ(v) = 1/m. Initialize the consensus graph S = Pm v=1 µ(v)G(v). For simplicity, we search both α and β in the range [0.05, 0.1, 0.5, 1, 2, 5, 10]. Motivated by (Nie, Cai, and Li 2017), we initialize the initial graphs G(v) by selecting 20-nearest neighbors among raw data.