DS_EVENT

DataScientia Workshop Meeting (17-18/04)

On April 17-18 the DataScientia community launch event was held in Trento (organised by the Department of Information Engineering and Computer Science and Department of Sociology and Social Research). DataScientia is a Citizen Science project initiated by UniTrento, which involves partner universities across the world, from Europe, to Asia, to Africa and South America. Among the involved institutions there are ECIU partners DCU and AAU. The community aims at involving people who are interested in contributing to, learning about, and influencing the future Data-driven Artificial Intelligence (AI) technological evolution, for themselves and for society.

Know More
image

CyCAT H2020 project for Algorithmic Transparency

We are happy to announce that the our group is now a partner in the CyCAT H2020 Project for next two years.  The main goal of the CyCAT is to address the emerging research problem of Algorithmic Transparency, and establish a network of researchers from European countries including Cyprus, England, Israel, Italy, and Scotland.

Know More
video_From language diversity to Bias

From language diversity to Bias (what about under-resourced languages)

A talk from Fausto Giunchiglia with Title “From language diversity to Bias (what about under-resourced languages)”. In this talk, Fausto provides a general overview of the problem of under-resources languages, and of how we could deal with this problem by uniformly handling the diversity which is intrinsic in any language. Trento April 5th, 2020.
(duration: 59min).

Watch here
PTMNTS

When Social Networks Become ‘Diversity-Aware’ Research Platforms

PYMNTS interviews Dr. Fausto Giunchiglia, project coordinator for  WeNet project.
Bigger, better, faster — those are often the guiding principles for innovation, disruption and change in the digital and mobile worlds. What about smarter, and more diverse?

Read the article
wenetgrp

WENET - DIVERSITY IS KEY FOR NEW, AI-BASED SOCIAL MEDIA PLATFORM -11th February 2019 Trento, Italy

The Horizon 2020 funded WeNet research project officially started in January with a kick-off meeting in snowy Trento, Italy. WeNet is a multidisciplinary project– using computer science, sociology and engineering – that will create a platform that enables people to support each other in a way that transcends geographical and cultural backgrounds. Diversity is a key enabler of the WeNet platform. It will use machine-learning algorithms to build user-profiles based on behavior and other key factors. The profiles will then be matched to positively exploit the diversity of users, allowing them to help one another. By connecting people from diverse backgrounds and skillsets, WeNet will enable them to interact and incentivize them to assist one another in ways that would not formerly have been possible. The WeNet platform will be the basis of a series of studies within universities worldwide with diverse student populations to improve students' quality of life inside and outside the academic environment, taking into consideration ethical and privacy guidelines. WeNet will start with a number of Smart University pilots to exploit diversity case studies. These will be extended by further engaging 5 new pilot sites that will be found via the Open Call funding system. For ensuring the long-term sustainability of the research infrastructure, a non-profit organization will be created, which will be in charge of managing the infrastructure and managing the community. This highly skilled consortium, led by the University of Trento, is composed of 15 partners from Europe and Israel, India, Mongolia, Paraguay, Mexico and China. The complementarity of the consortium partners is one of the greatest added values of WeNet, which includes highly rated universities and SMEs.

Download press release
43950651_717638168613938_5777471448301961216_n

Abdelhakim Fraihat's visit to Tshwane University of Technology- Pretoria (2018-10-01)

Abdelhakim Fraihat from University of Trento visited Tshwane University of Technology and conducted a five days research training workshop programme on the topic "Computational Linguistics: Lexical Semantics and Natural Language Processing."

Program info
36256122_1708468129229933_4807639503260876800_n

Fausto's Talk- Pretoria (2018-06-27)

On 27th of June 2018, Prof. Fausto Giunchiglia from University of Trento is addressing a keynote on the topic: "One world - seven thousand languages". 

Program info
20180526_214741

International workshop on Advances on Machine learning - Zhuhai (2018-05-26)

International workshop on Advances on Machine learning, Zhuhai (China), May 26, 2018. Fausto Giunchiglia presents the results of the 2017 Smart University project in Trento.

DSC_0181

DataScientia workshop - Trento (2016-05-11)

The aim of the workshop is to bring together professionals and practitioners from academic, business, and government communities from all over the world interested in exploring how diversity affects data and knowledge, and discuss about applications, supporting methodologies and tools. The event is the follow-up of the first two editions held in Bangalore and Trento in the context of the ITPAR project between Italy and India. The initiative is now attracting attention from a growing number of institutions world-wide.The workshop is organized into sessions, each of them addressing one specific topic. Sessions will be interactive to discuss and agree about joint activities to be done together in the future, and possible joint projects.The event is co-funded by the ITPAR project (Italy-India), the Department of Science and Technology (India), and the DISI department of the University of Trento (Italy) and it is sponsored by the “Ministero degli Affari Esteri e della Cooperazione Internazionale” (Italy).

Program info
IMG_9176

DataScientia workshop - Bangalore (2013-11-18)

We organized our first KiDF workshop with a motto "Unity and Diversity".  Professors from various leading universities in India, China, Mongolia and Trento participated in the event. 

_DSC3282

Not-for-profit cause launched for studying language diversity using artificial intelligence - Trento, Italy, 04 February, 2019.

Led by the University of Trento in Northern Italy, in collaboration with six other partner institutions from Europe, China, India, Mongolia, Mexico, South Africa and South America, researchers from these institutions have announced the launch of the new initiative DataScientia aiming at studying and connecting the diversity and unity of the world’s languages, with a special focus of under-resourced languages. The consortium of founding members includes Italy’s University of Trento, the National University of Mongolia, India’s Amrita University, China’s Jilin University, Paraguay’s Catholic University “Nuestra Señora de la Asunción”, South Africa’s Tshwane University of Technology, and UK’s Heriot-Watt University. The consortium will be working on several world languages including Italian, Gaelic, Spanish, Tamil, Malayalam, Hindi, African, Guarani and Mongolian. The workshop detailing this multi-community and multi-disciplinary project, held on January 29-30, 2019 at Trento, Italy, covered various enabling disciplines and technologies ranging from Artificial Intelligence and Computational Linguistics to linguistics, philosophy and social sciences.The DataScientia consortium has already set up projects for the development of various systems, among them: the Universal Knowledge Core (UKC) which allows the integration of the lexicons of the languages used to describe the world as perceived by the various cultures and languages, an engine for corpora collection, the SCROLL framework for multilingual Natural Language Processing, the data integration platform Digital University and a service provision platform based on behavioural patterns called Smart University.Prof. Fausto Giunchiglia from the University of Trento, who started this initiative, said: “This is a very important event which lays the basis for a new generation of research initiatives. The development of a single multilingual resource which integrates the world languages, its evolution in time, and its use for the study of the diversity and unity of languages, can be done meaningfully only with the participation of the people and Universities coming from and representing the relevant cultures. To succeed, we must develop this initiative as a win-win project where everybody brings their own culture and gets back the knowledge of how it relates to that of the others.” The consortium aim is to create a non-profit organisation which will ensure the long term sustainability of the project. While doing this, it will also engage and partner with many other University and groups for furthering this unity through diversity initiative.

Download press release
disi-unitn

NSURL 2019 The First Workshop on NLP Solutions for Under Resourced Languages Trento, 11-12 September 2019

The workshop aims to draw the attention of researchers in the under resourced languages  communities and encourage them to cooperate and intensify efforts to provide solutions and  resources for such under resourced languages. To achieve this goal, the workshop offers the possibility for researchers to organize competitions for solving NLP tasks. These tasks are usually provided with training and test data by the task organizers to allow participating solutions to be evaluated and compared with other solutions. The number of shared tasks for this year are 8 tasks for 5 languages (http://nsurl.org/tasks/).  Task description papers and the participating solutions including the performance evaluations will be published in the workshopproceedings. In addition, we invite submissions on topics that include but are not limited to  the following:

Low level NLP solutions:
    POS-tagging
    Tokenization
    Named entity  recognition
    Morphological analysis
    Chunking
    Disambiguation

Resources for low level NLP:

    Dictionaries
    Annotated data
    Corpora

Important dates:
Submission deadline: 30 June 2019
Notification of acceptance: 22 July 2019
Camera-ready paper due: 12 August 2019
Workshop dates: 11,12 September 2019
Submission pagehttps://easychair.org/conferences/?conf=nsurl2019

Website: http://nsurl.org/

article many languages

The many languages missing from the internet

English and a handful of other languages dominate the internet, but this is leaving indigenous cultures without a voice online. Now they are fighting to get their own languages on the web.

Imagine your favourite social media platform does not let you post in English. Now think of a keyboard that won’t allow you to type in your own words. You would have two options: either switch to another language or remain digitally silent.

This is the reality for most people that speak indigenous languages and dialects.

There are nearly 7,000 languages and dialects in the world, yet only 7% are reflected in published online material, according to Whose knowledge?, a campaign that aims to make visible the knowledge of marginalised communities online.

unnamed.jpg1

Workshop on NLP Solutions for Under Resourced Languages

The 3rd International Conference on Natural Language and Speech Processing (www.icnlsp.org/icnlsp2019), was held on 12-13 September 2019, at the Department of Computer Science and Information Engineering of the University of Trento, Trento. The conference was an opportunity and a forum for researchers and students to exchange ideas and discuss research and trends in the field of Natural Language Processing and Speech Processing. The total number of submitted papers was 46 papers, where 20 of them with an acceptance rate of 43 % were accepted. The presented papers covered various topics dealing with both speech and text:  building resources, text summarization, spoken language understanding, etc. The program included two keynotes. The first one, entitled “Detecting the fake news before they were even written”, was presented by Dr. Preslav Nakov from Qatar Computing Research Institute (QCRI),Qatar.  The second keynote “One world – seven thousand languages” was  presented by Prof.  Fausto Giunchiglia from University of Trento, Italy. The conference was preceded by the workshop on NLP Solutions for Under Resourced Languages (NSURL). The workshop is concerned with solving NLP problems for low-resourced languages.

Screen Shot 2018-06-19 at 10.35.04 AM

AfriKiD workshop - Pretoria (2017-09)

We organized our first AfriKiD workshop with a motto "Unity and Diversity".  The event was organized by the Tshwane University of Technology. 

Program info
2016Trento

DataScientia workshop - Trento (2017-03-29)

The aim of the workshop is to bring together professionals and practitioners from academic, business, and government communities from all over the world interested in exploring how diversity affects data and knowledge, and discuss applications, supporting methodologies and tools. The event is the follow-up of the first three editions, held in Bangalore and Trento, in the context of the ITPAR project between Italy and India. The initiative is now attracting attention from a growing number of institutions world-wide. The workshop is organized into sessions, each of them addressing one specific topic. Sessions will be interactive to discuss and agree about joint activities to be done together in the future, and possible joint projects.We organized our third KiDF workshop with a motto "Unity and Diversity".  Professors from various leading universities in India, China, Mongolia and Trento participated in the event. 

Program info
DSC_0181

UKC workshop - Changchun (2014-10-12)

This talk will present our going work, started more than 6 years ago, whose goal is to enable data integration among data collected in different cities, about different but related topics, in different languages, thus enabling cross-language data management. The first part will present the theoretical foundations. The second part will discuss the concrete steps towards making this project concrete, namely the creation of a no-profit organization whose goal will be to enable interoperability among the different sites. 
Program info