Call for participation: Shared Task TIAD-2017 Translation Inference Across Dictionarieshttps://tiad2017.wordpress.com
Overview Various methods and techniques have been explored in the past in the aim of automatically generating new bilingual (and multilingual) dictionaries from existing ones, for instance using one (or more) language(s) as a pivot between two other source and target languages. However, such efforts were usually conducted on different types of datasets and evaluated in different ways, making it difficult to compare due to the different experimental setups and evaluation metrics. TIAD-2017 is launched with the intention of offering quality lexical resources for a coherent experiment that enables reliable validation of results and solid comparison of methods and techniques used for the automatic generation of translations across languages. This initiative aims also to stimulate and enhance further research on the topic. It will make use of cross-lingual lexicographic data of K Dictionaries (KD), which will serve also to validate the results along with human assessment. The systems developed by participants and their results will be presented at a workshop that will be held as part of the first Language, Data and Knowledge conference in Galway, Ireland, on 18-20 June 2017 (http://ldk2017.org). The papers describing the participant systems will be published on CEUR-WS (http://ceur-ws.org).
Task definition The objective of the task is to indirectly generate translations for three language pairs, based on already known translations among eight languages in 14 bilingual dictionaries, involving four possible paths – all from German to Brazilian Portuguese – that feature between 1 to 4 pivot languages. The test dataset consists of 100 randomly-selected German dictionary entries with their translations into a second language, and recursively exploring further translations in chained-up dictionaries – including up to 1,035 entries with 1,948 translation equivalents in the largest language pair that is provided. Besides the headwords and translations, the data includes information about parts of speech, subject domains and synonyms, as well as examples of usage and their translations.
The following language pairs are provided for the four paths: (a) German > English > Portuguese (b) German > Japanese > Spanish > Portuguese (c) German > Danish > French > Spanish > Portuguese (d) German > Dutch > Spanish > Danish > French > Portuguese
Also included are four Portuguese > German datasets, for closing the loop in each path, to help with the validation of the results.
The three new language pairs that should be generated are: (1) German > Portuguese (2) Danish > Spanish (3) Dutch > French
Evaluation of the results of each system will be carried out against KD’s manually compiled dictionaries for these pairs from the Global Series and other resources, as well as by human translators.
Participants can contribute on either or both of the following tracks: (1) Systems that use only the KD data released for the task (2) Systems that exploit, in addition to the KD data, other freely available sources of background knowledge (e.g., lexical linked open data and parallel corpora) to improve performance
Beyond performance, participants are encouraged to consider the following issues in particular: • The role of the language family with respect to the newly generated pairs • The asymmetry of pairs, and how translation direction affects the results • The behavior of different parts-of-speech among different languages
Important Dates • 23.1.2017 – Call for participation / Test data released • 15.4.2017 – Submission of results by participants • 30.4.2017 – Evaluation of results communicated by organizers • 01.6.2017 – Submission of system description papers • 18.6.2017 – Workshop
Organizers • Jorge Gracia, Ontology Engineering Group, Universidad Politécnica de Madrid • Noam Ordan, K Dictionaries and The Arab Academic College of Education, Haifa • Ilan Kernerman, K Dictionaries, Tel Aviv
Review Committee To be announced on February 1.
Terms and Website A full description of TIAD-2017 and its binding terms and regulations is available on the website: https://tiad2017.wordpress.com/.
Contact Noam Ordan: noam at kdictionaries.com -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 4911 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20170125/fec7a471/attachment.txt>