Timezone: »

Is Automated Topic Model Evaluation Broken? The Incoherence of Coherence
Alexander Hoyle · Pranav Goel · Andrew Hian-Cheong · Denis Peskov · Jordan Boyd-Graber · Philip Resnik

@ None

Topic model evaluation, like evaluation of other unsupervised methods, can be contentious. However, the field has coalesced around automated estimates of topic coherence, which rely on the frequency of word co-occurrences in a reference corpus. Contemporary neural topic models surpass classical ones according to these metrics. At the same time, topic model evaluation suffers from a validation gap: automated coherence, developed for classical models, has not been validated using human experimentation for neural models. In addition, a meta-analysis of topic modeling literature reveals a substantial standardization gap in automated topic modeling benchmarks. To address the validation gap, we compare automated coherence with the two most widely accepted human judgment tasks: topic rating and word intrusion. To address the standardization gap, we systematically evaluate a dominant classical model and two state-of-the-art neural models on two commonly used datasets. Automated evaluations declare a winning model when corresponding human evaluations do not, calling into question the validity of fully automatic evaluations independent of human judgments.

Author Information

Alexander Hoyle (University of Maryland, College Park)
Pranav Goel (University of Maryland College Park (SSO))
Andrew Hian-Cheong (University of Maryland, College Park)
Denis Peskov (University of Maryland, College Park)
Jordan Boyd-Graber (University of Maryland)
Philip Resnik (University of Maryland)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors