Welcome to EVALITA 2018! EVALITA is the evaluation campaign of Natural Language Processing and Speech Tools for the Italian language. This year we celebrate 10 years of EVALITA and shared tasks covering the analysis of both written and spoken language with the aim of enhancing the development and dissemination of resources and technologies for Italian. EVALITA is an initiative of the Italian Association for Computational Linguistics (AILC, http://www.ai-lc.it/). It is supported by the Italian Association for Artificial Intelligence (AI*IA, http://www.aixia.it/) and by the Italian Association of Speech Science (AISV, http://www.aisv.it/). This volume collects the reports of the tasks organisers and of the participants to all of the EVALITA 2018 tasks. This year we helped to organize 10 tasks structured in four tracks: i.) Affect, Creativity and Style: Aspect-based Sentiment Analysis (ABSITA), Italian Emoji Prediction (ITAMoji), Irony Detection in Twitter (IronITA), Cross-Genre Gender Prediction (GxG); ii.) Dialogue Systems: itaLIan Speech acT labEliNg (iLISTEN), Italian DIALogue systems evaluation (IDIAL); iii.) Hate Speech: Automatic Misogyny Identification (AMI); Hate Speech Detection (HaSpeeDe); iv.) Semantics4AI: Solving language games (NLP4FUN), Spoken Utterances Guiding Chefs Assistant Robots (SUGAR). The volume opens with an overview to the campaign, in which we describe the tasks in more detail, provide figures on the participants, and, especially, highlight the innovations introduced in this year's edition. The abstract of Saif M. Mohammad's keynote with title “The Search for Emotions, Creativity, and Fairness in Language” is also included in this introductory part of the volume. The final workshop was held in Turin on the 12th and 13th of December 2018 as a co-located event of the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018, http://clic2018. di.unito.it/en/home-2/). The workshop has been an occasion for organizers and participants, from both academic institutions and companies, to disseminate their work and results and to share ideas through oral and poster presentations. This year the program also includes a panel on “The Future of Shared Tasks: data, evaluation, and technology transfer” as a moment of reflection after 10 years of EVALITA, also in light of the new regulations on privacy (i.e., the EU General Data Protection Regulation (GDPR)). A major innovation of this year is the best system award across all tasks. This wants to be a further incentive to involve students (both, undergraduates and Ph.D. students) and to push the boundaries of the state of the art further by “daring to experiment” rather than just winning. We thank all the people and institutions involved in the organisation of the tasks, and all the participants, who contributed to the success of the event. A special thank is due to AILC. Thanks are also due to AI*IA and AISV for endorsing EVALITA, to FBK and Manuela Speranza for making the web platform available once more for this edition (http://www.evalita.it), to our sponsors: CELI (https://www.celi.it/), Google Research, ELRA (http://elra.info/en/), and to Agenzia per l'Italia Digitale (AGID, https://www.agid.gov.it) for its endorsement. Last but not least, we heartily thank our invited speaker, Saif M. Mohammad from National Research Council Canada, for agreeing to share his expertise on key topics of EVALITA 2018 and to participate in our panel discussion, and all the other panelists: Franco Cutugno, (Universita` degli Studi di Napoli Federico II), Caterina Flick (Lawyer and coordinator of the Web & Media Committee of the Fe´de´ration Internationale des Femmes des Carrie`res Juridiques) and Malvina Nissim (Rijksuniversiteit Groningen).
Steering Committee Maria Anzovino, Universita` degli Studi di Milano-Bicocca, Italy Francesco Barbieri, Universitat Pompeu Fabra, Spain Pierpaolo Basile, Universita` degli Studi di Bari “A. Moro”, Italy Valerio Basile, Universita` degli Studi di Torino, Italy Cristina Bosco, Universita` degli Studi di Torino, Italy Francesca Chiusaroli, Universita` degli Studi di Macerata, Italy Alessandra Cignarella, Universita` degli Studi di Torino, Italy Danilo Croce, Universita` degli Studi di Roma “Tor Vergata”, Italy Francesco Cutugno, Universita` degli Studi di Napoli “Federico II”, Italy Marco De Gemmis, Universita` degli Studi di Bari “A. Moro”, Italy Felice Dell’Orletta, Istituto di Linguistica Computazionale “A. Zampolli”, CNR Pisa, Italy Maria Di Maro, Universita` degli Studi di Napoli “Federico II”, Italy Sara Falcone, Fondazione Bruno Kessler, Italy Elisabetta Fersini, Universita` degli Studi di Milano-Bicocca, Italy Simona Frenda, Universita` degli Studi di Torino, Italy Marco Guerini, Fondazione Bruno Kessler, Italy Bernardo Magnini, Fondazione Bruno Kessler, Italy Malvina Nissim, Rijksuniveristeit Groningen, The Netherlands Nicole Novielli, Universita` degli Studi di Bari “A. Moro”, Italy Antonio Origlia, Universita` degli Studi di Napoli “Federico II”, Italy Endang Wahyu Pamungkas, Universita` degli Studi di Torino, Italy Viviana Patti, Universita` degli Studi di Torino, Italy Fabio Poletto, Acmos, Torino, Italy Marco Polignano, Universita` degli Studi di Bari “A. Moro”, Italy Francesco Ronzano, Universitat Pompeu Fabra and Hospital del Mar Medical Research Center, Spain Paolo Rosso, PRHLT Research Center, Universitat Polite`cnica de Vale`ncia, Spain Manuela Sanguinetti, Universita` degli Studi di Torino, Italy Giovanni Semeraro, Universita` degli Studi di Bari “A. Moro”, Italy Lucia Siciliani, Universita` degli Studi di Bari “A. Moro”, Italy Maurizio Tesconi, Istituto di Informatica e Telematica, CNR Pisa, Italy Claudia Tortora, Universita` degli Studi di Napoli “L’Orientale”, Italy
Website