*** Apologies for cross-posting ***
Dear colleagues, We invite fluent speakers of either English, German or Czech to participate in an experiment on ranking machine translation outputs. The study aims to assess the ability of a new re-ranking approach to discriminate the best translation among a list of candidates. Annotators will be given a source sentence along with a reference translation to help them in their judgement. The list of MT outputs to rank is composed by the original translation (a.k.a 1-best) produced by a MT system (either SMT or NMT), opposed to translation candidates from the hypotheses space picked by our approach as potentially better alternatives. All translation candidates are displayed in random order. Important: - This experiment has been approved by the University of Sheffield's Ethics Committee. - All the data collected is anonymous and the annotations you make will be used in future research. - Your email is necessary to sign up and to enter a draw to win an Amazon voucher worth GBP 50. It will not be used for any other purposes. - One annotation task usually takes about 20 minutes to complete. - You can do as many annotation as you want, just click on the task names that are assigned to you. - We will evaluate all tasks after the end of the experiment. Only users with completed tasks and annotations conducted without suspicious or random behaviours will compete to win the Amazon voucher. Participants must be: - Over 18 years of age. - Fluent speakers of either English, German or Czech; both native and non-native speakers are welcome to participate.To participate, visit:
http://appraise.fredblain.org Please read the instructions carefully. Thank you, Fred Blain & Lucia Specia Department of Computer Science University of Sheffield, UK --www.dcs.shef.ac.uk/~frederic gpg-key: FC7C3BC0
signature.asc
Description: PGP signature
_______________________________________________ Mt-list site list Mt-list@eamt.org http://lists.eamt.org/mailman/listinfo/mt-list