Correlated Tag Learning in Topic Model

Shuangyin Li, Rong Pan Sun Yat-sen University, Yu Zhang, Qiang Yang Hong Kong University of Science and Technology
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:18-27, 2016.

Abstract

It is natural to expect that the documents in a corpus will be correlated, and these correlations are reflected by not only the words but also the observed tags in each document. Most previous works model this type of corpus, which are called the semi-structured corpus, without considering the correlations among the tags. In this work, we develop a Correlated Tag Learning (CTL) model for semi-structured corpora based on the topic model to enable the construction of the correlation graph among tags via a logistic normal participation process. For the inference of the CTL model, we devise a variational inference algorithm to approximate the posterior. In experiments, we visualize the tag correlation graph generated by the CTL model on the DBLP corpus and for the tasks of document retrieval and classification, the correlation graph among tags is helpful to improve the generalization performance compared with the state-of-the-art baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-li16a, title = {Correlated Tag Learning in Topic Model}, author = {Li, Shuangyin and University, Rong Pan Sun Yat-sen and Zhang, Yu and Technology, Qiang Yang Hong Kong University of Science and}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {18--27}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/li16a/li16a.pdf}, url = {https://proceedings.mlr.press/r14/li16a.html}, abstract = {It is natural to expect that the documents in a corpus will be correlated, and these correlations are reflected by not only the words but also the observed tags in each document. Most previous works model this type of corpus, which are called the semi-structured corpus, without considering the correlations among the tags. In this work, we develop a Correlated Tag Learning (CTL) model for semi-structured corpora based on the topic model to enable the construction of the correlation graph among tags via a logistic normal participation process. For the inference of the CTL model, we devise a variational inference algorithm to approximate the posterior. In experiments, we visualize the tag correlation graph generated by the CTL model on the DBLP corpus and for the tasks of document retrieval and classification, the correlation graph among tags is helpful to improve the generalization performance compared with the state-of-the-art baselines.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Correlated Tag Learning in Topic Model %A Shuangyin Li %A Rong Pan Sun Yat-sen University %A Yu Zhang %A Qiang Yang Hong Kong University of Science and Technology %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-li16a %I PMLR %P 18--27 %U https://proceedings.mlr.press/r14/li16a.html %V R14 %X It is natural to expect that the documents in a corpus will be correlated, and these correlations are reflected by not only the words but also the observed tags in each document. Most previous works model this type of corpus, which are called the semi-structured corpus, without considering the correlations among the tags. In this work, we develop a Correlated Tag Learning (CTL) model for semi-structured corpora based on the topic model to enable the construction of the correlation graph among tags via a logistic normal participation process. For the inference of the CTL model, we devise a variational inference algorithm to approximate the posterior. In experiments, we visualize the tag correlation graph generated by the CTL model on the DBLP corpus and for the tasks of document retrieval and classification, the correlation graph among tags is helpful to improve the generalization performance compared with the state-of-the-art baselines. %Z Reissued by PMLR on 04 October 2026.
APA
Li, S., University, R.P.S.Y., Zhang, Y. & Technology, Q.Y.H.K.U.o.S.a.. (2016). Correlated Tag Learning in Topic Model. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:18-27 Available from https://proceedings.mlr.press/r14/li16a.html. Reissued by PMLR on 04 October 2026.

Related Material