Datasets of Slovene and Croatian Moderated News Comments

摘要

This paper presents two large newly con-structed datasets of moderated news comments from two highly popular online news por-tals in the respective countries: the Slovene RTV MCC and the Croatian 24sata. The datasets are analyzed by performing manual annotation of the types of the content which have been deleted by moderators and by in-vestigating deletion trends among users and threads. Next, initial experiments on auto-matically detecting the deleted content in the datasets are presented. Both datasets are pub-lished in encrypted form, to enable others to perform experiments on detecting content to be deleted without revealing potentially inap-propriate content. Finally, the baseline classi-fication models trained on the non-encrypted datasets are disseminated as well to enable real-world use.