The Ministry of Science, Innovation and Universities is funding eleven research studies on the digitalization of society

The State Research Agency, under the Ministry of Science, Innovation and Universities, has justified the funding awarded to new research projects as part of its national R, D & I programme. On this occasion, the Universitat Oberta de Catalunya (UOC) has been awarded more than 600,000 euros to carry out eleven projects, thanks to a call for proposals focused on research challenges which contribute to the solution of social, economic and technological problems and to the generation of knowledge, thereby promoting research projects motivated by scientific curiosity, and encouraging the spread of quality scientific and technological knowledge that results in social improvements. In each of the two previous calls for proposals, seven UOC research projects received funding.

5G for the digitalization of industry

A project coordinated by Xavier Vilasajona, professor and research leader of the group Wireless Network (WiNE), affiliated to the Internet Interdisciplinary Institute (IN3), will study this technological development to promote the digitalization of industrial processes. Vilasajona points out, "It is possible to improve industrial processes with current software solutions and other alternatives from the field of telecommunications." The project Adaptation of 5G technologies and alternatives to support the digitalization of industry (SPOTS) is focused on the development of high-reliability wireless technology, in line with current developments promoted under the umbrella of the 5G era, for the consolidation of industrial digitalization; new robotic applications, for example. SPOTS hopes to demonstrate its results with functional prototypes, as well as use a network of companies in the region to illustrate the transfer to industry of the technologies developed.

Self-managed internet models

For some time, it has been possible to access the internet via community-owned data networks. These networks are an emerging model of decentralized, self-managed systems. Evidence of this is seen in the project guifi.net, an open and free telecommunications network based on a collaborative economy model through which users share their resources. The philosophy consists of providing other users with access to the network, especially favouring users based in areas where the coverage provided by commercial companies is insufficient. However, "infrastructure comprised of components provided by volunteers, as is the case with these community networks, have specific characteristics that affect their reliability and the quality of the service," as Joan Manuel Marquès points out. Marquès, a member of the group Internet Computing & Systems Optimization (ICSO), affiliated to the Internet Interdisciplinary Institute (IN3), leads this project, coordinated by AMICS, entitled Data assignment methods to improve the quality of service in community-owned systems, which aims to study and propose data assignment methodologies, make use of resources provided by users and improve access to and use of these community networks.

Urban resilience and hydroclimatic risks

The study of resilient cities has become popular in recent years. Urban resilience refers to urbs that adapt to major change, and are able to recover from crises and prevent these from becoming recurrent events, as in the case of crises related to climate change and natural disasters. "However, there is little critical research explaining the unexpected implications of resilience plans designed from the top down, especially for the most vulnerable urban populations, as well as the transformative capacities of bottom-to-top initiatives," say Isabel Ruiz Mallen, Ramón y Cajal researcher, and Hug March, associate professor of the Faculty of Economics and Business at the UOC. The project RESCITIES, The political ecology of urban resilience to hydroclimatic events in Spain, led by Ruiz Mallen and March, members of the IN3 research group Urban Transformation and Global Change Laboratory (TURBA), will approach the analysis of resilience strategies applied to the cities of Barcelona and Seville from an urban political ecology perspective. Urban political ecology focuses on the analysis of socio-environmental change and sustainable city models through political action, affirming that whether or not these goals are achieved is determined by the relationships of those in power. RESCITIES will analyse how to promote citizen participation in the design of solutions for the challenges faced by these cities. The project will focus on an analysis of hydroclimatic risks and their implications for the inhabitants of Barcelona and Seville, and their relationship with climate change and socio-environmental change policies.

Social media networks to stimulate citizen participation among young people

"The internet is the channel most suited to the socio-cultural needs of young people, and the medium over which their parents have least control," says Daniel Aranda, deputy dean of the Faculty of Information and Communication Sciences at the UOC. In this context, the project Digital social education (ESD), led by Aranda who is a member of the research group Learning, Communications Media and Entertainment (GAME), will explore how the use of the internet and social media networks could promote the participation of young people in politics, boosting their engagement and their commitment to politics and society. The research project seeks to advance the digital social education of young people in their free time, in informal spaces and educational institutions, enhancing their skills and competences. One of the primary objectives of the project is to develop digital critical and analytical capacity in young people, thus boosting young citizen participation.

Using feedback received to promote learning

To ensure that their learning is effective, students receive teacher feedback on their assignments, addressing their results, concerns and queries. However, a number of different studies indicate that "students do not play an active role in the feedback process which is necessary to build their knowledge," says Teresa Guasch, dean of the Faculty of Psychology and Education Sciences at the UOC. The objective of the project ON-Feed: Involving students in the feedback dialogue for learning in virtual environments – led by Guasch, co-coordinator and researcher of the group Feed2Learn, together with professor Anna Espasa – is to analyse the reasons behind the students' lack of motivation in response to teacher feedback. Therefore, this project aims to enable students to use and interact with this feedback, and thus encourage student participation, reduce university drop-out rates and facilitate a decision-making tool for teaching staff to use with this feedback in virtual environments.

Corrections in the study of languages

The project Optimization of the online learning of languages through oral correction (TECSLA), is coordinated by Gisela Grañena, lecturer at the Centre for Modern Languages of the Faculty of Arts and Humanities and lead researcher of the technology-mediated acquisition of second languages research laboratory (TechSLA Lab). Grañena is working with Laia Canals, Aleksandra Malicka and Yucel Yilmaz studying the effectiveness of corrective feedback – that is, oral correction and comments by teaching staff with respect to oral tasks performed by students – in the learning of languages. Taking into consideration that, according to Grañena, "immediate feedback in response to an error tends to be most effective" and that "the online learning conditions in a context like the UOC only allow for deferred rather than immediate feedback", TECSLA will analyse feedback known as deferred immediate feedback, that is, asynchronous feedback which imitates the immediacy of a synchronous context. The project aims to examine other factors that could influence the effectiveness of the feedback and interact with the synchronization or asynchronization, such as the type of correction (more or less explicit feedback), the mode (written or oral), and individual cognitive differences between students. The research team hopes to be able to contribute to virtual language learning theory and improvements in course design.

A study of society as a networked ecosystem

The study of complex systems, that is, those formed by a large number of interacting particles, has become an incipient sphere of cross-cutting research. It is important, not only for understanding particular physical phenomena, but because of its influence on disciplines such as ecology, biology, economy, sociology or engineering. This is the case for the discipline referred to as computational human ecology, that is, the functioning of society understood as an ecosystem that conditions, for example, the characteristics of traffic in large cities. "Fundamental aspects are lacking in our understanding of how networked systems are born, grow and function," says Javier Borge. The project Hybrid patterns in uni- and multi-layered complex networks: theory and applications (HYPAT), coordinated by Borge, lead researcher of the IN3 group Complex Systems (CoSIN3), will analyse patterns that determine phenomena related to the human population. The project aims to design methodologies to facilitate understanding of the networked functioning of society, as well as study other related complex systems.

Legal framework for the use of artificial intelligence

Artificial intelligence offers social benefits in many sectors, but the lack of an adequate legal framework can have implications for our fundamental rights and freedoms. "Abuses of privacy and data protection rights can arise, and this leads to a lack of legal security," say Raquel Xalabarder, professor and dean of the Faculty of Law and Political Science at the UOC, and Miquel Peguera, a lecturer of the same faculty. The project Artificial Intelligence and the Law (LAI), led by Xalabarder and Peguera as coordinator and researcher, respectively, both members of the group Internet Law (DDI), aims to present legislative and political proposals that address the risks engendered by the inappropriate use of this technology. One of the project's aims is to analyse the limitations of the existing legal framework for the use of text mining systems, data analytics powered by artificial intelligence and other ways of processing and exploiting data sets for socially beneficial purposes, and recommend appropriate legislative measures. The research team will study the risks of using artificial intelligence from a number of different perspectives including intellectual property and privacy protection, especially in the case of automated decision making.

Improve the automated digital classification of emotions

Digital technologies such as computer vision applied to object recognition, surveillance, facial recognition and medical diagnostic systems are booming. The current dominant paradigm is the use of deep learning techniques, which in many cases obtain results close to human capacities. But these methods can only offer a single response in relation to the task at hand: for example, determine who a face belongs to, or whether cancerous cells are detectable in a given image. Real-life applications need more information, in particular some type of measure of the degree of uncertainty, in other words, how sure is the system, and what is the reasoning behind a particular interpretation. "For example, it is not enough that a medical application, having analysed a scanned image, suggests that a patient could have cancer. We need to know how reliable the diagnosis is, and on what the system is basing its diagnosis, in visual form if possible," says researcher David Masip. Masip, dean of the Doctoral School and lecturer of the Faculty of Computer Science, Multimedia and Telecommunications at the UOC, and Ágata Lapedriza, both members of the Scene, Understanding and Artificial Intelligence Lab (SUNAI), lead the project Deep learning models considering explainability and uncertainty: applications to emotional perception (BECAUsE). This research project seeks to define the uncertainty of these models, to determine the accuracy of an algorithm that automatically determines to what a given image or sound corresponds, as well as to understand the reasoning behind its decisions and be able to monitor it. The project will also analyse applications that allow machines to have an automated perception of the emotional state of a person. "We have an infinite amount of unlabelled data in which emotions are apparent, in the audiovisual sphere, in cinema and, in general, on the internet. We are proposing the development of a protocol to enable algorithms to use this vast amount of data, in a monitored way, to learn our emotions and understand which parts of the image – such as the facial expression, the context or the pose – contribute to emotional perception," says David Masip.