Fine-Grained Opinion Corpus


Priberam Fine-Grained Opinion Corpus

This page provides links and information about the Priberam Fine-Grained Opinion Mining Corpus, a Portuguese (MPQA-like) fine-grained dependency opinion mining corpus, which was used in [1] as a validation set. The corpus contains a 80 news documents extracted from the Priberam Compressive Summarization Corpus, having a total of 1226 sentences, annotated with 828 opinions (direct-subjectives) and respective agents, targets, polarities and intensities.

Download

Download here the Priberam Fine-Grained Opinion Mining Corpus.

You are probably interested in a Portuguese translation of the MPQA Subjective lexicon that we also used in [1]: the  Priberam Subjectivity Lexicon for Portuguese.

License

The corpus is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. For the legal licensing terms, please see the LICENSE.txt file in the archive. You can find a human-readable summary of the license (which is not a substitute of the license) here.

References

If you use this corpus in your research, please cite the following paper:

[1] Mariana S. C. Almeida, Claudia Pinto, Helena Figueira, Pedro Mendes and André F. T. Martins. 2015. Aligning Opinions: Cross-Lingual Opinion Mining with Dependencies", In Annual Meeting of the Association for Computational Linguistics (ACL).

Acknowledgements

This work was partially supported by the EU/FEDER programme, QREN/POR Lisboa (Portugal), under the Intelligo project (contract 2012/24803) and by a FCT grants UID/EEA/50008/2013 and PTDC/EEI-SII/2312/2012.