Topic Modeling on Health Journals With Regularized Variational Inference

Authors: Robert Giaquinto, Arindam Banerjee

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our results show significant improvements over competing topic models particularly after regularization, and highlight the DAP model s unique ability to capture common journeys shared by different authors. ... Section 5 introduces the evaluation dataset and procedure. Section 6 shares the results of the experiments.
Researcher Affiliation Academia Robert Giaquinto, Arindam Banerjee Dept of Computer Science & Engineering University of Minnesota, Twin Cities {smit7982@umn.edu, banerjee@cs.umn.edu}
Pseudocode No No structured pseudocode or algorithm blocks are provided. The generative process of the model is described in prose.
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets No The Caring Bridge (CB) dataset is mentioned: 'The full dataset includes 13.1 million journals written by approximately half a million authors between 2006 and 2016. From the CB dataset we draw an evaluation dataset consisting of journals written by authors who posted, on average, at least twice a month over a one year period.' However, no link, DOI, repository, or citation for public access to this specific dataset is provided.
Dataset Splits No The paper states: 'Journals are split into training and test sets with 90% of each author s journals (N = 103, 018) for training and 10% (N = 11, 728) for testing.' It also mentions '10-fold cross validation'. However, a separate validation split (e.g., for hyperparameter tuning) is not explicitly mentioned.
Hardware Specification No The paper mentions 'University of Minnesota Supercomputing Institute (MSI) for technical support' in the acknowledgments, but no specific hardware details such as GPU/CPU models, processor types, or memory used for experiments are provided.
Software Dependencies No No specific software dependencies with version numbers (e.g., library names with versions) are mentioned.
Experiment Setup Yes Following the approach of others, we simply fix the number of topics at 25 for all models. The number of personas learned by the DAP model is fixed at 15.