The paper presents the MediaBubble Dataset. Developing the dataset, our primary aim is to fill the gap of political topic detection dataset for Hungarian, a low density language. The dataset contains 1000 political articles appeared in the Hungarian on-line media in political topics on the major news portals between 26.04.2017 and 29.04.2017. The dataset contains the topics and topic assignments created by 3 annotators. In addition, the dataset is initiated as a crowdsourcing dataset. It means that although the dataset is publicly available, in order to download it, a dedicated amount of annotations has to be conducted as a contribution to research.
@InProceedings{GRAD-GYENGE18.21, author = {Lászó Grad-Gyenge and Linda Andersson}, title = {The MediaBubble Dataset}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Claudia Soria and Laurent Besacier and Laurette Pretorius}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-22-1}, language = {english} }
Summary of the paper