The 5th evaluation campaign EVALITA 2016 was organized along the following selected tasks:

  • ArtiPhon – Articulatory Phone Recognition
  • FactA – Event Factuality Annotation
  • NEEL-IT – Named Entity rEcognition and Linking in Italian Tweets
  • PoSTWITA – POS tagging for Italian Social Media Texts
  • QA­4FAQ – Question Answering for Frequently Asked Questions
  • SENTIPOLC – SENTIment POLarity Classification

New Challenge

In addition, we launched a new challenge organized by IBM Italy:

The EVALITA 2016 final workshop was held in Naples on December the 7th, 2016, as a satellite event of the 3rd Italian Conference on Computational Linguistics – CLiC-it 2016.

EVALITA 2016 is an initiative of AILC (Associazione Italiana di Linguistica Computazionale).


Proceedings are available on the CEUR open access platform and on the aAccademia University Press website.

Final Ranking

The final ranking of the IBM Watson Challenge is the following:

  1. Walter Basile and Valerio Basile: Stockle: feel the market.
  2. Giovanni Moretti: MARTIN: Monitoring and Analysing Real-time Tweets in Italian Natural language
  3. Gianni Barlacchi, Alessandro Moschitti, Andrea Mariani, Emanuele Rossinelli: Appetitoso: Search the dish, find the restaurant!

Congratulations to the winners!

Follow EVALITA 2016 on Twitter and Facebook and use the hashtag #EVALITA2016 to disseminate the initiative!!

Read the Storify story on the final workshop.

Read the journal paper on EVALITA 2016 published on IJCoL – Italian Journal of Computational Linguistics vol. 3, n. 1: “EVALITA Goes Social: Tasks, Data, and Community at the 2016 Edition” by Pierpaolo Basile, Malvina Nissim, Viviana Patti, Rachele Sprugnoli and Francesco Cutugno

Data are available on github: https://github.com/evalita2016/data

We thank IBM-Italy for supporting the IBM Watson Services Challenge.