Dall'8 al 9 febbraio 2024, presso Palazzo del Mediterraneo, l'Università degli studi di Napoli "L'Orientale" ospiterà circa 100 ricercatori in occasione del secondo meeting generale della COST-Action UniDive ("Universality, diversity and idiosyncrasy in language technology"), organizzato dall'UNIOR NLP Research Group. 

L'evento sarà preceduto il 7 febbraio dal Meeting del Work Group 1. 

La COST-Action UniDive, mira ad abbracciare la diversità inter- e intra-linguistica, cioè una diversità intesa sia in termini di differenze tra le lingue esistenti sia di varietà di fenomeni linguistici presenti all'interno di una stessa lingua. 

L’obiettivo di UniDive è quello di: (i) perseguire l’universalità delle terminologie e delle metodologie applicabili al Trattamento Automatico del Linguaggio (TAL), (ii) quantificare la diversità inter- e intra-linguistica, (iii) promuovere e coordinare lo sviluppo di risorse e strumenti linguistici orientati all’universalità e alla diversità. Riunisce circa 250 esperti in settori interdisciplinari (linguisti, linguisti computazionali, informatici, psicolinguisti e settori industriali) da quasi 40 paesi e rappresenta dozzine di lingue diverse.

L'evento all’Orientale rappresenta un'importante occasione di visibilità dell’Ateneo a livello internazionale, in cui lo scambio di conoscenze interdisciplinari costituirà un momento di importante confronto sulle attuali metodologie e sui progetti in corso riguardanti le diverse lingue rappresentate in UniDive. 

Si prevedono poster sessions, sessioni dedicate ai Working Groups (WG), interventi da parte di invited speaker, plenary talks e momenti di discussione. 

Il programma dettagliato dell'evento è disponibile qui

Dal 20 al 22 ottobre 2023 Raffaele Manna e Giulia Speranza, membri del gruppo di ricerca UNIOR NLP del Dipartimento di Studi Letterari, Linguistici e Comparati dell’Università degli studi di Napoli “L’Orientale” hanno partecipato come espositori alla undicesima edizione del Maker Faire di Roma (European edition).

L'evento, che si tiene ogni anno, rappresenta un’occasione unica per facilitare e raccontare l’innovazione in modo semplice e accessibile, connettendo le imprese, il mondo accademico, le persone e le idee. 

Nei tre giorni di fiera, nello spazio del padiglione 5 dedicato al Lab AIIS - CINI, il gruppo ha presentato il progetto ATHENA, sviluppato insieme a Maria Pia di Buono, frutto delle ricerche del gruppo nell’ambito della linguistica computazionale, che coniuga le tecnologie del linguaggio per il potenziamento e l’ottimizzazione di contenuti culturali in applicazioni innovative.

Our Research Group presented one contribution titled “Tailoring Terminological Resources to the Users’ Needs: A Corpus-based Study on Appositive Constructions in Italian and English” at the 1st International Conference on "Multilingual Digital Terminology Today: Design, representation formats and management systems (MDTT 2022)”, which took place on the 16th and 17th of June 2022 at the University of Padua, Italy.

On 18 January 2022, our PhD student Antonio Pascucci defended his PhD dissertation entitled "Does style characterize haters? An investigation on an Italian hate speech corpus". The defence thesis committee consisted of Professor Ruslan Mitkov (University of Wolverhampton, UK), Professor Paolo Rosso (Universitat Politècnica de València) and Professor Johanna Monti (Antonio's PhD supervisor). Congratulations Antonio and good luck!

On December 16th, Professor Johanna Monti and PhD students Antonio Pascucci and Raffaele Manna presented at the Aston Institute for Forensic Linguistics (Aston University)  their research titled "Style and falsification. The analysis of the statements of Brigate Rosse".

From October 21st to October 23rd, the PhD student Antonio Pascucci participated in the "Digital Forensics - Evidence Analysis via Intelligent Systems and Practices" COST Action 7th working group meeting, organized in Cádiz (Spain). On this occasion, Antonio presented his study on "Detecting stylistic features to identify hate speech spreaders on social media".

On October 20th 2021, the UNIOR NLP Research Group  participated in the seventh edition of the Graduate Conference, organised by the PhD students of the Department of Literary, Linguistic and Comparative Studies of the thirty-fifth cycle of the University of Naples "L'Orientale". The conference took place on 18, 19 and 20 October and the contributions provided by the participants focused on the theme "Islands and bridges. For a linguistic and literary topology of isolation'. The research group took part in the conference with a paper entitled 'Inclusion and accessibility. Technologies as a bridge between the past and the future", with the aim of reflecting on the use of technological tools to promote museum fruition in terms of accessibility. 

From 4th to 8th October 2021, the PhD student Antonio Pascucci took part in the Training school on Digital forensics: evidence analysis via intelligent systems and practices, organized in Budapest (Hungary) in the framework of the COST Action Digital Forensics Evidence Analysis via Intelligent Systems and Practices of which he has been a member since 2018. 

The UNIOR NLP research Group took part in the task force organised by Europeana.eu, entitled Europeana as a powerful platform for storytelling.The task force, coordinated by Beth Delay, took place entirely online, through monthly meetings and activities, from September 2020 to March 2021. The final product was the creation of a toolkit for cultural heritage institutions containing 7 tips for good digital storytelling. 

On September 24th, on the occasion of the 2nd International Conference of the European Association for Digital Humanities (EADH 2021), Antonio Pascucci presented the research titled “Collective intelligence as a resource to monitor environmental crimes. How to recognize an alert tweet”, a research carried out in cooperation between Antonio Pascucci, Wanda Punzi Zarino, Raffaele Manna, Vincenzo Simoniello, Annarita Magliacane and Prof. Johanna Monti in the framework of the C4E (Crowd for the environment) project.

The University of Naples “L’Orientale” took part in the Meet me tonight event – European Researchers’ Night (Notte Europea dei Ricercatori), which was organized on the 24th of September in Naples, Italy.  For this special event, the UniOR NLP Research Group organized the interactive laboratory “COLLECTIVE INTELLIGENCE: A RESOURCE FOR MONITORING THE ENVIRONMENT THROUGH SOCIAL MEDIA” aiming at presenting the C4E project, whose scope is to develop an innovative framework for the detection and monitoring of environmental crimes on the Italian territory.

Furthermore, the University of Naples “L’Orientale” presented to the public the Apps created by the students during the L’ORIENTALE APPLE FOUNDATION PROGRAM.

From 10th to 19th September 2021, the UNIOR NLP research Group took part in the Summer School Tran(s)missions: how multimedia shapes and reforms interdisciplinary research in the field of Italian studies and visual culture, organised by the University of Roma Tre and the Virtual Humanities Lab of the Brown University. During the summer school the meaning of interdisciplinarity and multimedia was explored by bringing together different types of studies from the fields of Italianistics, Digital Humanities, digital culture studies and research methods focused on the materiality of the object. On the occasion of the final project, a physical exhibition was held at Spazio Taverna in Rome, and a virtual one within the New Art City platform, which hosted the Parallel Disruptions project.

Our Research Group presented one contributions at the Workshop "Forensic linguistics between scientific research and legal practice" hosted in the congress of the Italian Linguistics Society - SLI 2020 (Società Linguistica Italiana) , which took place virtually on the 9th September 2021.

Raffaele Manna, Antonio Pascucci and Johanna Monti presented their research titled “La misurazione stilistica della Falsificazione. I comunicati delle Brigate Rosse”.  This study aims to show the possibilities of linguistic investigation in the forensic field using the techniques of computational stylometry. Taking as example the Communiqués sent by the Red Brigades during the Moro kidnapping, it will be demonstrated a posteriori to what extent the fake communiqué (communiqué 7) can be said to be such and what are the linguistic features that characterize it unlike the original ones.

The UNIOR NLP research group participated in the Digital storytelling festival that took place online from 1st to 31st March 2021, with the aim of encouraging people to write digital stories using openly licensed cultural heritage collections. The group participated in the festival with a digital story about some cat mummies dating back to the 1st century B.C., preserved in the Museo Scerrato in Naples, Italy. The story was published on the online platform Medium and it was given the people choice award.

On 11, 12 and 13 November 2020, the UNIOR NLP Research Group participated in the conference organised by Europeana.eu, entitled Crisis, Change and Culture. In the session Love stories - digital storytelling experiences we can't stop thinking about, the research group proposed its own contribution on the achievements of digital storytelling in the cultural heritage sector, reflecting on the cultural videogame Past for future, set in the National Archaeological Museum of Taranto (MArTA).

C4E @CLiC-It2020  - The first results of a research carried out in the framework of the C4E (Crowd 4 Environment) project will be presented in March 2021 at The Seventh Italian Conference on Computational Linguistics (CLiC-iT 2020).  The Crowd 4 Environment (C4E) project aims at developing an innovative framework for the detection and monitoring of illegal spills, such as illegal landfills, micro-dumps or illegal releases in surface waters, and the organisation of subsequent monitoring actions on site.

The contribution of our research group is focused on the development of methods devised to assist the organisations responsible for the environmental monitoring and protection, and in particular on the acquisition of useful information through social media monitoring through NLP techniques. We built a model that is able to discriminate between alert and no-alert tweets on the basis of an annotated subsection of UNIOR Eye corpus related to environmental crimes: Here you can find a first map with preliminary results on alerts concerning illegal fires posted on Twitter between 2013-2020: 

The ItaGLam Corpus is composed of 3,429 tweets published by 213 Italian GLAMs, according to six hashtags used on Twitter, during the lockdown period caused by the spread of the COVID-19 virus (#artyouready, #emptymuseum, #museitaliani, #museichiusiuseiaperti, #laculturanonsiferma, #laculturaincasa). The corpus is annotated according to four communication categories:

• Sharing Experience (SE): tweets that share an experience, an opinion, or one’s own feeling; 

• Promoting Participation (PP): tweets that require some kind of activities, be it in real life or on-line; 

• Interacting with the Community (ItC): tweets through which the Institutions create and foster their own community by directly interacting with users; 

• Promoting Informing (PI): tweets that promote or inform the users about activities, exhibitions, or about any sort of information on the museum; 

This study stems from a previous work carried out during the early days of the pandemic available at the following link https://arxiv.org/abs/2005.10527.

This research, carried out together with Francesca Chiusaroli and Maria Laura Pierucci from the university of Macerata, focuses on the analysis of the language used by Italian university students on the web during the lockdown due to COVID-19 pandemics. The data used were collected from the “Spotted” accounts on Instagram, which represent a very informal channel of communication, and as such can provide important insights on the spontaneous writing of young people. We collected a corpus of 3,.865 posts converted with an OCR tool from pictures to strings of text. To showcase the importance of the corpus, we decided to describe the linguistic changes that happened after the beginning of the lockdown in Italy. This was done by analyzing two sub-corpora: in the first one we collected the posts written between the 10th of January and the 7th of March, while the second was made up of posts written from the 8th of March to the 4th of May (i.e. during the lockdown). 

Argentina Rescigno, graduate student, presents at AMTA2020 Virtual Workshop on the Impact of Machine Translation (# iMpacT2020) a work together with Johanna Monti, Andy Way, Eva Vanmassenhove of the ADAPT research center entitled " Gender bias in Neural Machine Translation" 

The work is the result of a collaboration with the prestigious Irish research center during a period of stay of Argentina for an Erasmus + mobility. The work is part of a particularly interesting line of research that addresses the issue of gender translation in machine translation and presents the results of an evaluation conducted on Google Translate, DeepL and Bing Microsoft Translator regarding natural gender translation providing statistics on the frequency of feminine, masculine and neutral forms in the translation of a list of personality adjectives, names of professions and epicenes. The evaluation concerns the English → Spanish, English → Italian and English → French language pairs. Here you can find the link of the presentation: https://www.youtube.com/watch?v=lQz1_gpY5GE .

The ELEXIS project organized a shared task on the task of monolingual word sense alignment across dictionaries as part of the GLOBALEX 2020 – Linked Lexicography workshop at the 12th Language Resources and Evaluation Conference (LREC 2020) taking place on Tuesday, May 12 2020 in Marseille (France).

Monolingual word sense alignment is a challenging task of finding matching senses between two dictionary entries and will play a crucial role in the development of new lexical resources. In addition, this task presents a challenging combination of NLP, semantic textual similarity and reasoning in order to find the best alignment across a group of senses.

You can find our results here.  

Our Research Group took part in the Festival about Science and Technology “Futuro Remoto 2019 - Essere 4.0” at the Science Centre of Città della Scienza, Naples, from 21 to 24 November 2019.

Annalisa Raffone presented her Lab on Digital Storytelling and gaming for the acquisition of English idioms in classroom.  Storytelling & Gaming Lab

Antonio Pascucci presented “Il Mago della Ghigliottina”, a linguistic game based on Computational Linguistics and AI, capable of resolving the last step of the TV game-show “La Ghigliottina”. Il Mago della Ghigliottina

Raffaele Manna presented a system based on AI to spot cases of cyberbullying in texts, with the aim of shading lights on the negative effects on the victims. AI for Cyberbullying and Psychological Distress Detection

Giulia Speranza presented the Apps developed during the Program “L’Orientale” Apple Foundation which periodically takes place at the University of Naples “L’Orientale”, giving the chance to students of humanities to learn coding. L'Orientale Apple Foundation Program for Smart Communities

The UniOR NLP Research Group took part in the 12th Edition of the International Conference - Innovation in Language Learning (ILL 2019) organized by Pixel (International Education and Training Institution) with the aim to promote transnational cooperation in the field of Innovation for Education.

The Conference took place in Florence on November 14-15, 2019. 

Our Research Group contributed to the advances in the field of Second Language Acquisition through innovative approaches with two papers:

Annalisa Raffone presented her study on Gaming for Language Learning and MWE assimilation with E-voluzione s.r.l. and Prof. Johanna Monti entitled "LightVerb-Quest: An Adventure Storytelling Game to Foster Second Language Acquisition and Verbal Multi-Word Expressions Assimilation".

Annalisa Raffone presented her study on Digital Storytelling for Second Language Acquisition with Prof. Johanna Monti entitled "A Digital Storytelling Laboratory to foster Second Language Acquisition in Higher Education: Students' Perspectives and Reflections".

Link paper gaming here

Link paper Digital Storytelling here

The UniOR NLP Research Group took part in the 16th Edition of the International Conference - Cognition and Exploratory Learning in the Digital Age (CELDA 2019) organized by IADIS (International Association for Development of the Information Society) with the aim to address the main issues concerned with evolving learning processes and supporting pedagogies and applications in the digital age.

The Conference took place at the University of Cagliari on November 7-9, 2019. 

Our Research Group contributed to the educational arena with one paper:

Annalisa Raffone presented her study on Digital Storytelling for Second Language Acquisition in Higher Education with Prof. Johanna Monti entitled "Becoming Storytellers: Improving ESL Students' Academic Engagement and 21st Century Skills through Interactive Digital Storytelling".

Link to the Conference Program here

Our PhD Students Annalisa Raffone and Antonio Pascucci took part in  CILGI Conference (Quarto Convegno Internazionale di Linguistica e Glottodidattica Italiana), held in Campobasso on 26-28 September 2019. Annalisa and Antonio held two laboratories:

Antonio: Giochi linguistici e nuove tecnologie.

Annalisa:  EvoQuest: un sistema per creare giochi didattici di gruppo

Our PhD student Giulia Speranza took part in the Summer School “I progetti Wikimedia per le Istituzioni Culturali” held at the Wikimedia Italia Foundation in Milan on 9-13 September 2019.  

During the Summer School, participants followed seminars and lectures about Wikipedia and its related projects such as Wikidata, Wikisource and WikiCommons and how the so-called GLAM (Galleries, Libraries, Archives, Museums) can collaborate and sharing their data to the community. Wikidata, in particular, was described as the wikipedian solution to the Linked Data Principles on the basis of the Semantic Web. Several hours were also dedicated to the SPARQL query language to semantically explore data. 

During the week several projects in support of Cultural Heritage were illustrated such as Wiki Loves Monuments. In addition to lectures, also a practical session in the form of an Editathon (marathon of editing) was organized, where participants edited a Wikipedia page about a masterpiece of Leonardo Da Vinci collection stored in the Museo della Scienza e della Tecnica in Milan, extracting info from material kindly offered by the museum itself. 

Summer School Programme can be found here

September 3rd 2019, our PhD Student Antonio Pascucci took part in the 4th Workshop on Emotion and Sentiment in Social and Expressive Media: Media and Arts for Inclusive, Fair and Reflective Societies (ESSEM), co-located with the 8th International Conference on Affective Computing & Intelligent Interaction (ACII 2019), held in Cambridge, United Kingdom.

The ESSEM workshop is focused on models and applications that can impact on monitoring, analyzing inequalities and on supporting the implementation of countermeasures to foster inclusion and fairness. Inequalities are indeed increasingly spreading phenomena in combination with the pervasive of social media, which may play a key role with respect to exclusion of youngers (cyberbullying), women (misogyny), and immigrants (hate speech), just to cite few related challenges for the ESSEM community.

On this occasion, Antonio presented a talk entitled Computational Stylometry and Machine Learning for Gender and Age Detection in Cyberbullying Texts, carried out in cooperation with Professor Johanna Monti and Vincenzo Masucci (of the Expert System Corp.)

Workshop Programme can be found here

Our PhD Student Antonio Pascucci took part in the 1st International Summer School on Computational Forensics (SuCoFo2019) held in Lillehammer (Norway) on 26–30 August 2019. The Summer School has been organized in the context of the COST Action CA17124 “Digital forensics: evidence analysis via intelligent systems and practices”.

During the summer school, participants followed lectures on Digital Forensics and the week spent in Lillehammer gave the participants a unique opportunity to hear influential keynote speakers as well as benefit from interaction and discussions with international experts within the field. The course covered both theoretical and practical aspects of computational forensics and pattern recognition problems, as well as examples of their successful commercialization.

During the summer school, Antonio gave a talk about Computational Stylometry and the importance of Computational Linguistics and Authorship Attribution in Digital Forensics.

On this occasion, Antonio presented a talk entitled Computational Stylometry and Machine Learning for Gender and Age Detection in Cyberbullying Texts, carried out in cooperation with Professor Johanna Monti and Vincenzo Masucci (of the Expert System Corp.)

Antonio Pascucci, PhD Student of our research group, took part in the Codificare, Analizzare, Diffondere: Le Digital Humanities nei progetti di ricerca summer school, funded by the AIPI (Associazione Internazionale Professori di Italiano with the contribution of the AIUCD (Associazione per l'Informatica Umanistica e la Cultura Digitale). 

The summer school took place in Verona on 16-19 July 2019.

During the Summer School, participants followed lectures on Computational Stylometry, Natural Language Processing, Semantic Analysis, Stylistic Analysis, Optical Character Recognition methods, Cartography, Text Encoding, HTML Encode, and how tho disseminate the research on the Web.