A constrained optimization approach for cross-domain emotion distribution learning

Xiaorui QIN*, Yufu CHEN, Yanghui RAO, Haoran XIE, Man Leung WONG, Fu Lee WANG

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

Abstract

Emotion distribution learning aims to annotate unlabeled instances with a set of emotion categories and their strengths. Non-negative Matrix Tri-Factorization (NMTF) introduces an association matrix between document clusters and word clusters to help the domain adaptation task in emotion distribution learning. Nevertheless, many prior cross-domain emotion distribution learning methods had two major deficiencies. First, they hypothesize that there is a one-to-one correspondence between document clusters and emotion labels. In their experiments, the number of document clusters depends on the number of labels. Second, the prior work does not endow models with adequate constraints. In the real scenario of cross-domain emotion distribution learning, there are potential constraints that may improve the performance of such models. In order to address these problems, we propose a constrained optimization approach based on NMTF for cross-domain emotion distribution learning. In our model, the relationship between document clusters and emotion labels is not always one-to-one. A novel content-based constraint is also endowed based on the hypothesis that documents belonging to the same clusters must have similar content. We solve the optimization problem by an alternately iterative algorithm and show the proof of convergence. Experiments on 12 real-world cross-domain emotion distribution learning tasks validate the effectiveness of our method.
Original languageEnglish
Article number107160
JournalKnowledge-Based Systems
Volume227
Early online date27 May 2021
DOIs
Publication statusPublished - 5 Sep 2021

Bibliographical note

Funding Information:
The research described in this article has been supported by HKIBS Research Program, China Grant ( HCRG-201-002 ), Faculty Research, China Grant ( DB21A9 ) and LEO Dr David P. Chan Institute of Data Science of Lingnan University, Hong Kong , and a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China ( UGC/FDS16/E01/19 ).

Publisher Copyright:
© 2021 Elsevier B.V.

Keywords

  • Content-based constraint
  • Domain adaptation
  • Emotion distribution learning
  • Non-negative Matrix Tri-Factorization

Fingerprint

Dive into the research topics of 'A constrained optimization approach for cross-domain emotion distribution learning'. Together they form a unique fingerprint.

Cite this