DMCA
Exploring Topic Coherence over many models and many topics
| Citations: | 22 - 0 self |
BibTeX
@MISC{Stevens_exploringtopic,
author = {Keith Stevens},
title = {Exploring Topic Coherence over many models and many topics},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We apply two new automated semantic evaluations to three distinct latent topic models. Both metrics have been shown to align with human evaluations and provide a balance between internal measures of information gain and comparisons to human ratings of coherent topics. We improve upon the measures by introducing new aggregate measures that allows for comparing complete topic models. We further compare the automated measures to other metrics for topic models, comparison to manually crafted semantic tests and document classification. Our experiments reveal that LDA and LSA each have different strengths; LDA best learns descriptive topics while LSA is best at creating a compact semantic representation of documents and words in a corpus. 1
Keyphrases
many model topic coherence many topic coherent topic internal measure semantic test document classification complete topic model human rating human evaluation topic model new aggregate measure different strength semantic evaluation automated measure learns descriptive topic compact semantic representation information gain distinct latent topic model
