From: Ruben V. <rub...@ug...> - 2014-12-01 15:48:33
|
Hi Pablo, Maybe first some context: I was asked by the Dutch DBpedia chapter to perform this work. However, not everybody (including me) was aware of this existing code. When I reported back, Dimitris send me a pointer to it. > Do you think this could improve on the current Gender extractor that Max and I created? Yeah; the ideas seem very similar; the main difference being my use of scores (which are not necessary for English, but are for other languages like Dutch where some pronouns are homonyms of other words). > We'd love to have it improved. Why don't you send a pull request over there? So because I didn't know it existed; and because I never contributed to a Scale project. But the first obstacle has been overcome, and the second one should be fairly easy :-) It would probably be a good idea to cross-check first; i.e., are the results consistent with Wikidata and/or the existing algorithm. Is your result dataset available somewhere for 2014? Mine is here: https://github.com/RubenVerborgh/DBpediaGenderResults Best, Ruben |