{"id":291,"date":"2019-10-09T06:15:34","date_gmt":"2019-10-09T06:15:34","guid":{"rendered":"http:\/\/sag.art.uniroma2.it\/NL4AI\/?page_id=291"},"modified":"2020-06-19T11:39:41","modified_gmt":"2020-06-19T11:39:41","slug":"workshop-program-2018","status":"publish","type":"page","link":"http:\/\/sag.art.uniroma2.it\/NL4AI\/workshop-program-2018\/","title":{"rendered":"Workshop Program (NL4AI 2018)"},"content":{"rendered":"

News
\n<\/strong>The Workshop proceedings have been published: http:\/\/ceur-ws.org\/Vol-2244\/<\/a><\/p>\n

\n

Nov 22, 11.00 -11.10<\/p>\n

Introduction to the workshop<\/strong><\/h3>\n
\n

Nov 22, 11.10 – 12.10<\/p>\n

Invited Speaker<\/strong><\/h3>\n

Malvina Nissim<\/a> <\/strong><\/h4>\n

Title<\/strong><\/span>: Who can do author profiling?<\/span><\/span><\/p>\n

Abstract<\/strong>: <\/span><\/p>\n

State-of-the-art performance on author profiling (e.g., gender and age) for English in social media is around 80%, or even higher. I will present systems that yield those results, but I will also question the reliability of such figures.<\/div>\n
<\/div>\n
Specifically, I will show that the same reasons why these models work are also the cause of serious performance drops when we change, even moderately, domain or topic. If we create conditions to bypass such powerful but at the same time limiting clues, we might be able to identify features that are indeed more relevant for profiling, and we will build more portable systems.<\/div>\n
<\/div>\n
I will describe experiments in this direction, exploring whether portability can be stretched even into a cross-language setting. I will also compare these flexible models to human performance in quite an interesting way!<\/div>\n
<\/div>\n
\n
\n<\/div>\n

Nov 22, 12.10 – 12.40<\/p>\n

Invited Speaker<\/strong><\/h3>\n

Giuseppe Attardi<\/a><\/strong><\/h4>\n

Title<\/span><\/strong>: The GARR Container Platform for AI <\/span><\/p>\n

 <\/p>\n

 <\/p>\n

\n

Nov 22, 12.40 – 13.00<\/p>\n

Poster Booster<\/h3>\n

Transfer Learning for Industrial Applications of Named Entity Recognition [pdf]<\/a><\/strong>
\nLingzhen Chen, Alessandro Moschitti, Giuseppe Castellucci, Andrea Favalli and Raniero Romagnoli<\/p>\n

Aspect Detection in Book Reviews: Experimentations [pdf]<\/a><\/strong>
\nJeanne Villaneau, Stefania Pecore and Farida Sa\u00efd<\/p>\n

Lexicon, Meaning Relations, and Semantic Networks [pdf]<\/a><\/strong>
\nPrakash Mondal<\/p>\n

A Sentence based System for Measuring Syntax Complexity using a Recurrent Deep Neural Network [pdf]<\/a><\/strong>
\nDaniele Schicchi, Giosue’ Lo Bosco and Giovanni Pilato<\/p>\n

Automatic Relation Extraction for Building Smart City Ecosystems using Dependency Parsing [pdf]<\/a><\/strong>
\nDaniel Braun, Anne Faber, Adrian Hernandez-Mendez and Florian Matthes<\/p>\n

 <\/p>\n

\n

Nov 22, 15.30 – 16.30<\/p>\n

Poster Session<\/h3>\n

 <\/p>\n

\n

 <\/p>\n

Nov 22, 16.30 – 17.30<\/p>\n

Oral Presentations – I<\/strong><\/h3>\n

Learning Representations for Biomedical Named Entity Recognition [pdf]<\/a><\/strong><\/span>
\nIvano Lauriola, Riccardo Sella, Fabio Aiolli, Alberto Lavelli and Fabio Rinaldi<\/em><\/p>\n

Evaluating Kernel-based Sentence Embeddings [pdf]<\/a><\/strong><\/span>
\nDanilo Croce, Simone Filice and Roberto Basili<\/em><\/p>\n

Cross-Attentive CNN for Question Answering Sentence Selection<\/strong><\/span>
\nAlessio Gravina, Federico Rossetto, Silvia Severini and Giuseppe Attardi<\/em><\/p>\n

 <\/p>\n

\n

Nov 23, 9.00 – 10.00<\/p>\n

Oral Presentations – II<\/strong><\/h3>\n

Multi-Task Learning in Deep Neural Network for Sentiment Polarity and Irony classification [pdf]<\/a><\/strong><\/span>
\nLorenzo De Mattei, Andrea Cimino and Felice Dell’Orletta<\/em><\/p>\n

Task-oriented Conversational Agent Self-learning Based on Sentiment Analysis [pdf]<\/a><\/strong>
\nSerena Leggeri, Andrea Esposito and Luca Iocchi<\/em><\/p>\n

Exploiting Deep Neural Networks for Tweet-based Emoji Prediction [pdf]<\/a><\/strong><\/span>
\nAndrei Catalin Coman, Giacomo Zara, Yaroslav Nechaev, Gianni Barlacchi and Alessandro Moschitti<\/em><\/p>\n

 <\/p>\n

\n

Nov 23, 10.00 – 11.00<\/p>\n

Invited Speaker<\/strong><\/h3>\n

Oliviero Stock<\/a> <\/strong>has been at IRST (now called FBK-irst) since 1988 and has been its director from 1997 to 2001. His activity is in artificial intelligence, natural language processing, intelligent user interfaces, cognitive technologies, computational creativity. He is the author of over two hundred and sixty peer-reviewed papers and author or editor of twelve volumes, and has been a member of the editorial board of a dozen scientific journals. He has been the second President of AI*IA, Chairman of the European Coordinating Committee for Artificial Intelligence (ECCAI), President of the Association for Computational Linguistics, and is a EURAI and a AAAI Fellow.<\/p>\n

Title<\/strong>: Automated persuasion and linguistic creativity<\/p>\n

Abstract<\/strong>: Philosophers of language have taught us that at the basis of language production there is the intention to change the state of the world by intervening linguistically on other agents. Persuasion, being the process of influencing attitudes, beliefs, behaviors, mood of a target, is basically a matter of stronger emphasis. Argumentation is just one resource to persuasion, it has been studied since the times of Aristotle and now for quite some time in artificial intelligence.
\nIn modern times we have witnessed also resorting to nonargumentative means, belonging to what is called the peripheral route to persuasion. Automated intelligent persuasion of the latter sort is a research area close to producing usable results, both through creative production of language expressions, and through other forms of communication. The theme is sensitive and so it is equally important to address ethical acceptability of persuading machines.<\/span><\/p>\n

 <\/p>\n

\n

Nov 23, 11.30 – 12.10<\/p>\n

Oral Presentations – III<\/strong><\/h3>\n

Mapping Natural Langauge Labels to Structured Web Resources [pdf]<\/a><\/strong><\/p>\n

Valerio Basile, Elena Cabrio, Fabien Gandon and Debora Nozza<\/em><\/p>\n

Lexical Meaning Formal Representations Enhancing Lexicons and Associated Ontologies [pdf]<\/a><\/strong><\/span>
\nMaria Gritz<\/em><\/p>\n

 <\/p>\n

\n

Nov 23, 12.10 – 12.30<\/p>\n

Conclusion\/Panel<\/strong><\/h3>\n
\n

The schedule of the conference (and the workshops) can be found at the following link<\/a>.<\/p>\n

 <\/p>\n

The program of the previous edition of the NL4AI Workshop (held in Bari in 2017) is available at this link<\/a>.<\/p>\n

 <\/p>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

News The Workshop proceedings have been published: http:\/\/ceur-ws.org\/Vol-2244\/ Nov 22, 11.00 -11.10 Introduction to the workshop Nov 22, 11.10 – 12.10 Invited Speaker Malvina Nissim  Title: Who can do author profiling? Abstract:  State-of-the-art performance on author profiling (e.g., gender and age) for English in social media is around 80%, or even higher. I […]<\/p>\n","protected":false},"author":5,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"template-fullwidth.php","meta":[],"_links":{"self":[{"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/pages\/291"}],"collection":[{"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/users\/5"}],"replies":[{"embeddable":true,"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/comments?post=291"}],"version-history":[{"count":3,"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/pages\/291\/revisions"}],"predecessor-version":[{"id":349,"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/pages\/291\/revisions\/349"}],"wp:attachment":[{"href":"http:\/\/sag.art.uniroma2.it\/NL4AI\/wp-json\/wp\/v2\/media?parent=291"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}