Following the success of the previous editions, we are pleased to announce EVALITA 2020, the seventh evaluation campaign of Natural Language Processing and Speech tools for Italian.
The general objective of EVALITA (http://www.evalita.it <http://www.evalita.it/>) is to promote the development of language technologies for the Italian language, providing a shared framework where different systems and approaches can be evaluated in a consistent manner. EVALITA is an initiative of the Italian Association of Computational Linguistics <http://www.ai-lc.it/> (AILC) and is endorsed by the Italian Association for Artificial Intelligence <https://aixia.it/> (AIxIA) and the Italian Association for Speech Sciences <http://www.aisv.it/index.php> (AISV).
For the 2020 edition tasks are organized along the following tracks:
Affect, Hate, and Stance A <http://sag.art.uniroma2.it/absita/>TE_ABSITA <http://www.di.uniba.it/~swap/ate_absita/index.html> - Aspect Term Extraction and Aspect-Based Sentiment Analysis AMI <https://amievalita2020.github.io/> - Automatic Misogyny Identification SardiStance <http://www.di.unito.it/~tutreeb/sardistance-evalita2020/index.html> - Stance Detection HaSpeeDe <http://www.di.unito.it/~tutreeb/haspeede-evalita20/index.html> - Hate Speech Detection Creativity and Style CHANGE-IT <https://sites.google.com/view/change-it/home> - Style Transfer TAG-it <https://sites.google.com/view/tag-it-2020/> - Topic, Age and Gender Prediction Semantics and Multimodality DANKMEMES <https://dankmemes2020.fileli.unipi.it/> - Multimodal Artefacts Recognition CONcreTEXT <https://lablita.github.io/CONcreTEXT/> - Concreteness in Context Ghigliottin-AI <https://ghigliottin-ai.github.io/> - Evaluating Artificial Players for the Language Game “La Ghigliottina” PRELEARN <https://sites.google.com/view/prelearn20/home> - Prerequisite Relation Learning Time and Diachrony DaDoEval <https://dhfbk.github.io/DaDoEval/> - Dating Documents DIACR-Ita <https://diacr-ita.github.io/DIACR-Ita/> - Diachronic Lexical Semantics New Challenges in Long-standing Tasks AcCompl-it <https://sites.google.com/view/accompl-it/home-page> - Complex Parsing KIPoS <http://www.di.unito.it/~tutreeb/kipos-evalita2020/index.html> - Part-of-speech Tagging on Spoken Language
As with other international evaluation campaigns, guidelines describing the individual tasks will be distributed among the participants. Participants to the tasks will also be provided with training data and will have the chance to test their systems according to the metrics and procedures to be used in the formal evaluation.
The results of the evaluation will be disseminated at the final workshop, which will be organised in Bologna as a co-located event of the 7th Italian Conference on Computational Linguistics - CLiC-it 2020 (November 30-December 2, 2020).
Due to the COVID-19 pandemic, the situation is still uncertain. EVALITA will coordinate with CLiC-it in the hope to see each other in Bologna this December. In the event that this is not possible, we will try to organise a virtual edition of Evalita 2020. In any case, we will publish the workshop proceedings. We have seen great efforts from the tasks organisers, and many tasks for this year are an absolute novelty for Evalita. We believe that it is fundamental for our research community to evaluate our progress on more traditional tasks as well as novel ones. We should not miss the opportunity to fuel the research on NLP and Computational Linguistics for Italian with your new and interesting ideas and approaches.
We encourage participation from both academic institutions and industrial organizations. Please visit the EVALITA web site <http://www.evalita.it/2020> where information about all tasks is currently available, and indicate your intention to participate and the specific task(s) you are interested in. You can do so by filling out the registration form at the following link <https://docs.google.com/forms/d/e/1FAIpQLSeaoRKynqNvpqh0rjgkADvv_-ZMNKPq-EI5FyyOqGAZ8zhLDQ/viewform?usp=pp_url&entry.1387948180=e.g.+UniTo&entry.1365700421=e.g.+University+of+Turin>, which will be available until September 4th 2020, when test sets will be made available by the task organizers.
http://www.evalita.it/2020/taskregistration <http://www.evalita.it/2020/taskregistration> Registration is compulsory to participate in the task(s) you selected, although it is not binding. In any case, independently of your registration, you can decide to withdraw from participating by not sending the system results during the evaluation stage.
IMPORTANT DATES: 29th May 2020: development data available to participants 4th September 2020: registration closes 4th - 24th September 2018: evaluation windows (tentative) and collection of participants results, as follows: 4rd - 11th Sept. 2020: Evaluation Window I - Tasks: ATE_ABSITA; HaSpeeDe; CHANGE-IT; DANKMEMES; DaDoEval 11th - 17th Sept. 2020: Evaluation Window II - Task: AMI; TAG-it; CONcreTEXT; DIACR-Ita; AcCompl-it 17th - 24rd Sept. 2020: Evaluation Window III - Tasks: SardiStance; Ghigliottin-AI; PRELEARN; KIPoS Definite dates will be published on the EVALITA website. 6th November 2020: technical reports due to organisers (camera-ready) 2nd-3rd December 2020: final workshop
Scientific coordination: Valerio Basile (University of Turin) Danilo Croce (University of Rome “Tor Vergata”) Maria Di Maro (University of Naples “Federico II”) Lucia Passaro (University of Pisa) Advisor: Nicole Novielli (University of Bari “A. Moro”)
FOLLOW US! Twitter: @EVALITAcampaign Facebook: https://www.facebook.com/evalita2020/ <https://www.facebook.com/evalita2020/> -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 48071 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20200406/ef379f20/attachment.txt>