Anúncios
Prova de Doutoramento da Aluna Mojgan Hashemian
28 maio 2021, 12:19 • Sandra Espírito Santo
PROVA DE DOUTORAMENTO
Candidata : Mojgan Hashemian
TÍTULO DA TESE: “Persuasive Social Agents using Social Power Dynamics”
LOCAL DA PROVA:
Link de Zoom: https://videoconf-colibri.zoom.us/j/86392765845
DATA: 09/06/2021 HORA: 14:00H
Orientador: Rui Filipe Fernandes Prada
Co-orientador: Pedro Alexandre Simões dos Santos
Thesis Abstract:
Social Power, the potential for social influence, is a pervasive social process in human-human interactions. Despite its acknowledged role in social interaction, little attention has been paid to this phenomenon in human-agent interaction. One prominent example of social agents that have been of interest these days is the evolution of robotic agents. Further, with respect to robotic agents, limited studies have addressed social power in human-robot interactions. Recent advances on Social Robotics raise the question of whether a social robot can be used as a persuasive agent. To date, different attempts have been performed using several approaches to tackle this research question. In a nutshell, the objective of the present work is empowering intelligent agents with social power dynamics to develop more persuasive, believable social agents. On the one hand, having proper decision making facing social power increases the rationality of agents and enhances their believability. On the other hand, understanding different sources of social power boosts their persuasibility. In this thesis, we report the results of our recent advancements for this objective and draw suggestions for future directions. In sum, we first approached the problem by conceptualizing social power inspired by a theory proposed by French and Raven. Then we performed different user-studies to investigate how different bases of social power contribute to the persuasiveness of social robots. In Study 1, within an adversarial setting, we programmed two robots with expert and reward social power. As a specific instance of reward, we used social rewards (telling a joke by the robot). In the second study, a single robot was used to persuade the user using two different strategies (reward and coercion) comparing to a control condition. Finally, in the last study (Study 3), the robot used one power strategy (reward) but with different levels in different conditions, comparing to two control conditions with/without the presence of any robot. Overall, the results of our user studies endorse that social power (in particular, reward, coercion, and expert bases) endows persuasiveness to social robots. And different persuasive strategies could be perceived and be preferred differently considering users’ profiles or the study conditions.
Orientador: Rui Filipe Fernandes Prada
Co-orientador: Pedro Alexandre Simões dos Santos
Thesis Abstract:
Social Power, the potential for social influence, is a pervasive social process in human-human interactions. Despite its acknowledged role in social interaction, little attention has been paid to this phenomenon in human-agent interaction. One prominent example of social agents that have been of interest these days is the evolution of robotic agents. Further, with respect to robotic agents, limited studies have addressed social power in human-robot interactions. Recent advances on Social Robotics raise the question of whether a social robot can be used as a persuasive agent. To date, different attempts have been performed using several approaches to tackle this research question. In a nutshell, the objective of the present work is empowering intelligent agents with social power dynamics to develop more persuasive, believable social agents. On the one hand, having proper decision making facing social power increases the rationality of agents and enhances their believability. On the other hand, understanding different sources of social power boosts their persuasibility. In this thesis, we report the results of our recent advancements for this objective and draw suggestions for future directions. In sum, we first approached the problem by conceptualizing social power inspired by a theory proposed by French and Raven. Then we performed different user-studies to investigate how different bases of social power contribute to the persuasiveness of social robots. In Study 1, within an adversarial setting, we programmed two robots with expert and reward social power. As a specific instance of reward, we used social rewards (telling a joke by the robot). In the second study, a single robot was used to persuade the user using two different strategies (reward and coercion) comparing to a control condition. Finally, in the last study (Study 3), the robot used one power strategy (reward) but with different levels in different conditions, comparing to two control conditions with/without the presence of any robot. Overall, the results of our user studies endorse that social power (in particular, reward, coercion, and expert bases) endows persuasiveness to social robots. And different persuasive strategies could be perceived and be preferred differently considering users’ profiles or the study conditions.
Provas de CAT do aluno Nuno Filipe Verdelho Trindade
18 maio 2021, 15:13 • Sandra Espírito Santo
Thesis Title
Immersive Analytics in AEC
Nuno Filipe Verdelho Trindade Provas no dia 28 de maio de 2021, às 14h30.
Link de Zoom : https://videoconf-colibri.zoom.us/j/87823020554?pwd=VkpVZ3p2TDZlcWlmODBXZXM2NlM4UT09
Supervisor: Professor Alfredo Manuel dos Santos Ferreira Júnior
Co- Supervisor: Professor Daniel Jorge Viegas Gonçalves, e Professor Sérgio Bruno Martins de Oliveira
Thesis Abstract
Immersive analytics can be broadly described as the analysis of data using immersive technologies. Such technologies can enhance the understanding of the data, improve analytical reasoning, and support decision-making. The transition from traditional visual data analytics to immersive analytics has multiple aspects. These aspects include the use of immersive collaborative environments, multisensory stimuli, natural interfaces, and enhanced contextualization of data in the physical world using situated analytics. In this document, we present a research proposal and corresponding work plan concerning the use of immersive analytics in the Architecture, Construction, and Engineering (AEC) area. We discuss the state-of-the-art in immersive analytics and how AEC can benefit from its generalized adoption. In that sense, we delve into the innovative use of visualization and interaction technologies and their integration in companies and institutions' workflows. Our objective is to discover and analyze the tangible advantages of immersive analysis in real-world scenarios. We propose to explore aspects like the performance, insightfulness, abstraction, and contextualization potential of immersive analytics compared with traditional data analytics.
Thesis Abstract
Immersive analytics can be broadly described as the analysis of data using immersive technologies. Such technologies can enhance the understanding of the data, improve analytical reasoning, and support decision-making. The transition from traditional visual data analytics to immersive analytics has multiple aspects. These aspects include the use of immersive collaborative environments, multisensory stimuli, natural interfaces, and enhanced contextualization of data in the physical world using situated analytics. In this document, we present a research proposal and corresponding work plan concerning the use of immersive analytics in the Architecture, Construction, and Engineering (AEC) area. We discuss the state-of-the-art in immersive analytics and how AEC can benefit from its generalized adoption. In that sense, we delve into the innovative use of visualization and interaction technologies and their integration in companies and institutions' workflows. Our objective is to discover and analyze the tangible advantages of immersive analysis in real-world scenarios. We propose to explore aspects like the performance, insightfulness, abstraction, and contextualization potential of immersive analytics compared with traditional data analytics.
Provas de CAT do aluno João Manuel Fernandes Cardoso
12 maio 2021, 13:59 • Sandra Espírito Santo
Thesis Title
Addressing Research Data Management with Semantic Technology
João Manuel Fernandes CardosoProvas no dia 17 de Junho de 2021, às 18H30
Link de Zoom: https://videoconf-colibri.zoom.us/j/87562564571?pwd=UUVCbjVNZlM3dzVRU08xeXZRSnBQdz09
Supervisors: Professor José Borbinha, e Professor Daniel Faria
Thesis Abstract
The role of data in the scientific process is becoming increasingly relevant, bringing new challenges concerning the management of the data lifecycle. This is especially relevant concerning the growing number of requirements by research communities and funding bodies for the application of open access policies and of the FAIR data principles (findability, accessibility, interoperability, and reusability). In relation to the related concerns, the concept of data management plan emerged as a fundamental artefact to support research data management as a process. The data management plan is a document that is typically either a part of project proposals, detailing how research data is to be managed throughout its lifecycle, or of project outputs, detailing how that was done and how that management must proceed after the project’s end. However, data formats and tools for managing these types of artefacts are still limited, resulting in inefficient manual practices. This research work explores the potential of using machine-actionable data management plans as an hypothesis for the automation of research data management activities. To that effect, it will explore the possibility of establishing an ontological specification that enables formal representations of data management plans, which in turn will allow the automation of the related management activities. This work also will demonstrate how those results will apply to real scenarios of bioinformatics (namely at a national network with BioData.pt and at an international network with ELIXIR).
Supervisors: Professor José Borbinha, e Professor Daniel Faria
Thesis Abstract
The role of data in the scientific process is becoming increasingly relevant, bringing new challenges concerning the management of the data lifecycle. This is especially relevant concerning the growing number of requirements by research communities and funding bodies for the application of open access policies and of the FAIR data principles (findability, accessibility, interoperability, and reusability). In relation to the related concerns, the concept of data management plan emerged as a fundamental artefact to support research data management as a process. The data management plan is a document that is typically either a part of project proposals, detailing how research data is to be managed throughout its lifecycle, or of project outputs, detailing how that was done and how that management must proceed after the project’s end. However, data formats and tools for managing these types of artefacts are still limited, resulting in inefficient manual practices. This research work explores the potential of using machine-actionable data management plans as an hypothesis for the automation of research data management activities. To that effect, it will explore the possibility of establishing an ontological specification that enables formal representations of data management plans, which in turn will allow the automation of the related management activities. This work also will demonstrate how those results will apply to real scenarios of bioinformatics (namely at a national network with BioData.pt and at an international network with ELIXIR).
Provas de CAT do aluno Bruno Alexandre Barbosa Fragoso
10 maio 2021, 12:16 • Sandra Espírito Santo
Thesis Title:
Alignment between permanent and temporary governance structures in enterprise digital transformation
Bruno Alexandre Barbosa FragosoProva no dia 01 de junho de 2021, às 9h00
Link de Zoom : https://videoconf-colibri.zoom.us/j/82268684195?pwd=WjFXcGp5MXY4SHlldXlRc1ZkZi9xUT09
Supervisors: Professor André Ferreira Ferrão Couto e Vasconcelos, e Professor José Luís Brinquete Borbinha
Thesis Abstract:
- ** Alignment between permanent and temporary governance structures in enterprise digital transformation programs ** The increasing pace that organizational Information Systems (IS) are required to adapt and evolve requires new approaches on how transformation processes are governed over the daily operations. The governance roles and relationships associated with creating new governance structures result in increased complexity and promote inefficiency and loss of value if not correctly dealt. The hypothesis to explore in this work is that an Enterprise Architecture (EA) continuous governance alignment method can handle such complexity. While transformation processes (specifically digital transformation), projects are commonly addressed by enterprises as instruments for efficient strategy implementation. A project promotes enterprise transformation as a temporary endeavor with assigned governance roles to actors. A regular occurrence is that these scenarios encourage double assignments by setting temporary roles to actors who already have permanent business roles in the organization where the project will be scoped. And the more complex and heterogenous the transformation scopes are, the more that is a verified fact. Unbalanced authority roles, inaccurate reporting responsibilities, unclear ownership, and difficult accountability are just some of the events directly impacting the overall strategy output. The existing literature supports these findings, particularly noted in projects with a strong focus on using digital technology as a driver of change. And, for those in particular, state of the art evidence that still lacks a coherent solution that copes with the associated complexity and heterogeneity that underlies the scope of the transformations. As proposed under this research, such a solution can be based on the continuous alignment between permanent and temporary governance structures – by assessing, measuring, and aligning the assignments and authorities associated with business roles and their actors. This research aims to provide a conceptual framework for continuous alignment between permanent and temporary governance structures through Enterprise Architecture. With such aim, the following set of activities are put forward in this research: i) definition of a metamodel for the assessment of permanent and temporary governance structures; ii) analysis on the adequacy of existent structural and behavioral elements to evidence and represent misalignments, and iii) definition of architectural viewpoints constraints. Finally, it is also presented a case study demonstrating the misalignment in an enterprise's governance structure and how the research activities are expected to contribute to the development of an Enterprise Architecture continuous governance alignment method.