General Tensor Spectral Co-clustering for Higher-Order Data

Authors: Tao Wu, Austin R. Benson, David F. Gleich

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We use experiments on both synthetic and real-world problems to validate the effectiveness of our method. For the synthetic experiments, we devise a planted cluster model for tensors and show that GTSC has superior performance compared to other state-of-the-art clustering methods in recovering the planted clusters. In real-world tensor data experiments, we find that our GTSC framework identifies stop-words and semantically independent sets in n-gram tensors as well as worldwide and regional airlines and airports in a flight multiplex network.
Researcher Affiliation Academia Tao Wu Purdue University wu577@purdue.edu Austin R. Benson Stanford University arbenson@stanford.edu David F. Gleich Purdue University dgleich@purdue.edu
Pseudocode No The paper describes the algorithm using numbered steps within paragraphs, but it does not contain a structured pseudocode block or a clearly labeled algorithm section.
Open Source Code Yes Code and data for this paper are available at: https://github.com/wutao27/Gtensor SC
Open Datasets Yes Data were collected from http://openflights.org/data.html#route. English n-gram data were collected from http://www.ngrams.info/intro.asp and Chinese n-gram data were collected from https://books.google.com/ngrams.
Dataset Splits No The paper describes the generation of synthetic data and the real-world datasets used, but it does not specify explicit train/validation/test splits (e.g., percentages or sample counts) for reproducibility.
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, such as library names with version numbers, used to implement and run the experiments.
Experiment Setup Yes For our experiments, we found that high values (e.g., 0.95) of α do not impede convergence. We use α = 0.8 for all our experiments. For all of our experiments, we set the teleportation probability α = 0.8 and the minimum cluster size for splitting min_size = 2. We set the target conductance φ = 0.3.