I'm interested in language games, but can't see this format holding student attention. Has anyone on this board got language games that have been shown to work? ________________________________________ From: corpora-bounces at uib.no [corpora-bounces at uib.no] on behalf of Timothee Mickus [tmickus at atilf.fr] Sent: 21 May 2021 11:11 To: corpora at uib.no Subject: [Corpora-List] Game with a purpose on the distributional hypothesis
Hello,
[Apologies for cross-posting]
We are happy to present to you BlankCrack, available at the URL: https://blankcrack.atilf.fr/<https://www.google.com/url?q=https://blankcrack.atilf.fr/&sa=D&source=hangouts&ust=1621677914498000&usg=AFQjCNGij9YagGt0wr7_Zclsmi5hX8SB0w> . Please try our game out! We hope that you will enjoy playing and will spread the word.
This online game was created to study the validity of the distributional hypothesis. We are looking for words that share similar contexts and yet mean different things, as well as whether large language models like BERT and human annotators are comparable on such data.
Players are invited to either propose or solve riddles. Each riddle is constructed from two words: a player trying to solve one will be presented with sentences with some words blanked out, and will have to guess which of the two words correspond to these blanks.
Our game is available in five different languages:
We've included a number of features to make the game as enjoyable as possible: levels of difficulty, matches against friends... You can also log in through social media (Facebook, Twitter & Reddit).
Please have a look at the game, and if you like, please share the URL! If you have any feedback, feel free to reach us at the email address dblankcrack at atilf.fr<mailto:dblankcrack at atilf.fr> .
Thanks again,
Timothee MICKUS PhD Applicant at Atilf
_______________________________________________ UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora Corpora mailing list Corpora at uib.no https://mailman.uib.no/listinfo/corpora