Anchor-Free Correlated Topic Modeling: Identifiability and Algorithm

Authors: Kejun Huang, Xiao Fu, Nikolaos D. Sidiropoulos

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments using the TDT2 and Reuters-21578 corpus demonstrate that the proposed anchor-free approach exhibits very favorable performance (measured using coherence, similarity count, and clustering accuracy metrics) compared to the prior art.
Researcher Affiliation Academia Kejun Huang Xiao Fu Nicholas D. Sidiropoulos Department of Electrical and Computer Engineering University of Minnesota Minneapolis, MN 55455, USA huang663@umn.edu xfu@umn.edu nikos@ece.umn.edu
Pseudocode Yes Algorithm 1: Successive Projection Algorithm [6]... Algorithm 2: Anchor Free
Open Source Code No The paper describes its algorithms and experimental procedures but does not provide a link or explicit statement about making the source code for its method publicly available.
Open Datasets Yes In this section, we apply the proposed algorithm and the baselines to two popular text mining datasets, namely, the NIST Topic Detection and Tracking (TDT2) and the Reuters-21578 corpora.
Dataset Splits No The paper mentions using TDT2 and Reuters-21578 datasets for experiments but does not specify exact training, validation, or test split percentages, sample counts, or refer to predefined splits with citations for reproducibility of data partitioning.
Hardware Specification No The paper discusses runtime performance and computational complexity but does not specify the particular hardware (e.g., CPU, GPU models, or memory) used for running the experiments.
Software Dependencies No The paper mentions using 'well-established tools for eigen-decomposition' and 'the solver proposed in [18]' but does not provide specific version numbers for any software dependencies, libraries, or programming environments.
Experiment Setup No The paper describes data preprocessing steps (e.g., 'standard tf-idf data', 'normalized-cut weighted (NCW) [21] is applied') but does not provide specific experimental setup details such as hyperparameter values, optimizer settings, or explicit training configurations.