A Novel Document Generation Process for Topic Detection based on Hierarchical Latent Tree Models
This work addresses hierarchical topic detection for text analysis, offering a more efficient and interpretable approach, though it appears incremental as it builds on existing latent tree models.
The authors tackled the problem of hierarchical topic detection by proposing a novel document generation process using hierarchical latent tree models (HLTMs), which achieved drastically better model fit with fewer parameters and more meaningful topics compared to LDA-based methods.
We propose a novel document generation process based on hierarchical latent tree models (HLTMs) learned from data. An HLTM has a layer of observed word variables at the bottom and multiple layers of latent variables on top. For each document, we first sample values for the latent variables layer by layer via logic sampling, then draw relative frequencies for the words conditioned on the values of the latent variables, and finally generate words for the document using the relative word frequencies. The motivation for the work is to take word counts into consideration with HLTMs. In comparison with LDA-based hierarchical document generation processes, the new process achieves drastically better model fit with much fewer parameters. It also yields more meaningful topics and topic hierarchies. It is the new state-of-the-art for the hierarchical topic detection.