Improving the TENOR of Labeling: Re-evaluating Topic Models for Content Analysis
Topic models are a popular tool for understanding text collections, but their evaluation has
been a point of contention. Automated evaluation metrics such as coherence are often used …
been a point of contention. Automated evaluation metrics such as coherence are often used …
Labeled Interactive Topic Models
K Seelman, M Zhang, J Boyd-Graber - arXiv preprint arXiv:2311.09438, 2023 - arxiv.org
Topic models help users understand large document collections; however, topic models do
not always find the``right''topics. While classical probabilistic and anchor-based topic models …
not always find the``right''topics. While classical probabilistic and anchor-based topic models …
Studying the Effects of Collaboration in Interactive Theme Discovery Systems
APC Chen, D Srinivas, A Barry, M Seniw… - arXiv preprint arXiv …, 2024 - arxiv.org
NLP-assisted solutions have gained considerable traction to support qualitative data
analysis. However, there does not exist a unified evaluation framework that can account for …
analysis. However, there does not exist a unified evaluation framework that can account for …
Steerable Neural Topic Modeling
Q Fan, J Li - Proceedings of the 17th International Symposium on …, 2024 - dl.acm.org
This paper presents a Steerable Neural Topic Modeling (SNTM) technique. Unlike existing
techniques commonly relying on statistic-based topic models, we employ an autoencoder …
techniques commonly relying on statistic-based topic models, we employ an autoencoder …
Topic modelling for computational grounded theory
Z Fang - 2023 - wrap.warwick.ac.uk
The increasing amount of digital information has become a valuable source for research
whose sources of data are qualitative in nature. One commonly used method of content …
whose sources of data are qualitative in nature. One commonly used method of content …