Change search
ReferencesLink to record
Permanent link

Direct link
Semi-supervised latent variable models for sentence-level sentiment analysis
RISE, Swedish ICT, SICS. IAM.
Number of Authors: 2
2011 (English)Conference paper (Refereed)
Abstract [en]

We derive two variants of a semi-supervised model for fine-grained sentiment analysis. Both models leverage abundant natural supervision in the form of review ratings, as well as a small amount of manually crafted sentence labels, to learn sentence-level sentiment classifiers. The proposed model is a fusion of a fully supervised structured conditional model and its partially supervised counterpart. This allows for highly efficient estimation and inference algorithms with rich feature definitions. We describe the two variants as well as their component models and verify experimentally that both variants give significantly improved results for sentence-level sentiment analysis compared to all baselines.

Place, publisher, year, edition, pages
2011, 7.
National Category
Computer and Information Science
Identifiers
URN: urn:nbn:se:ri:diva-23853OAI: oai:DiVA.org:ri-23853DiVA: diva2:1042931
Conference
The 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies
Available from: 2016-10-31 Created: 2016-10-31

Open Access in DiVA

fulltext(384 kB)4 downloads
File information
File name FULLTEXT01.pdfFile size 384 kBChecksum SHA-512
49c687d875f9c5922a2cb9a2cc456d8c53a2f08e096388fba79760f72117a2c5fa62c5e31144e8655ac88a5523b16c7e5417877be1433ec29d82754e356e264e
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
McDonald, Ryan
By organisation
SICS
Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 4 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 1 hits
ReferencesLink to record
Permanent link

Direct link