Improving the TENOR of Labeling: Re-evaluating Topic Models for Content Analysis

arXiv:2401.16348v2 Announce Type: replace
Abstract: 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, however, their validity has been questioned for neural topic models (NTMs) and can overlook a models benefits in real world applications. To this end, we conduct the first evaluation of neural, supervised and classical topic models in an interactive task based setting. We combine topic models with …

abstract analysis arxiv automated cs.cl cs.cy cs.hc evaluation evaluation metrics labeling metrics popular text tool type understanding