Resources

Hungarian WordNet

Hungarian WordNet contains over 40,000 synsets and is fully linked to Princeton (English) WordNet 2.0 and 3.0.

It is available for download from https://github.com/dlt-rilmta/huwn. RDF version is available from https://github.com/dlt-rilmta/huwn.rdf

When referring to HuWN, please cite the following paper:

Miháltz, Márton, Csaba Hatvani, Judit Kuti, György Szarvas, János Csirik, Gábor Prószéky, Tamás Váradi: Methods and Results of the Hungarian WordNet Project. In: Proceedings of The Fourth Global WordNet Conference, Szeged, Hungary (2008), pp. 311–321. [pdf]

OpinHuBank

OpinHuBank is a human-annotated corpus to aid the research of opinion mining and sentiment analysis in Hungarian. It consists of 10,000 sentences containing person names (identified with the huntag NER tool) from major Hungarian news sites and blogs. Each entity occurrence was tagged by 5 human annotators for sentiment polarity in its sentence (neutral, positive or negative).

The corpus is available for download via the META-SHARE network, or you can also download it here.

Please cite the following paper when referencing OpinHuBank in your work:

Miháltz Márton: OpinHuBank: szabadon hozzáférhető annotált korpusz magyar nyelvű véleményelemzéshez. In Tanács Attila, Vincze Veronika (szerk.): IX. Magyar Számítógépes Nyelvészeti Konferencia (MSZNY 2013), SZTE, Szeged, 2013, pp. 343-345.

For more information, please read the paper (Hungarian).