[Topic-models] On author topic model
thibaut.thonet at irit.fr
Sun Jul 23 18:11:23 EDT 2017
You can also have a look at the following papers:
* Yang, M., & Hsu, W. H. (2016). HDPauthor: A New Hybrid Author-Topic
Model using Latent Dirichlet Allocation and Hierarchical Dirichlet
Processes. In Proceedings of the 25th International Conference
Companion on World Wide Web (pp. 619–624).
* Xuan, J., Lu, J., Zhang, G., Xu, R. Y. Da, & Luo, X. (2015).
Infinite Author Topic Model based on Mixed Gamma-Negative Binomial
Process. In Proceedings of the 2015 IEEE International Conference on
Data Mining (pp. 489–498). http://doi.org/10.1109/ICDM.2015.19
They both propose non-parametric author-topic models with hierarchical
priors, similar to what you described -- document-level topic
distributions' prior is based on author-specific topic distributions.
Le 21/07/2017 à 02:20, Eric Kang a écrit :
> Hi everyone,
> I have a question about the author-topic model. Is my understanding correct that the author-topic probabilities are "constant" across different documents? So if the same author writes multiple documents, the implied document-topic proportions would be the same between those documents?
> I thought perhaps another model might be to suppose that author-topic probabilities are a multinomial random variable (with a Dirichlet prior) that is sampled per document. In other words, each author is associated with author-specific Dirichlet distribution over topics, and for a particular document, a topic mixture is sampled from that Dirichlet distribution. And the inference problem would be to determine the topic-word probabilities, and the Dirichlet parameters for each author.
> Does this make sense? Is there existing work of this kind in the literature? Would this be interesting? Useful? Tractable?
> Any suggestions or guidance would be really appreciated.
> Thank you,
> Topic-models mailing list
> Topic-models at lists.cs.princeton.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Topic-models