Manuel Marques-Pita

Manuel Marques-Pita

Assistant Professor of Complex Systems

CICANT, Universidade Lusofona



Manuel Marques-Pita is an Assistant Professor of Complex Systems at CICANT, Universidade Lusófona (Portugal) and an Associate Researcher in the Complex Systems group at ISTAR (ISCTE-IUL) (Portugal). His research is on the interplay between individual and collective behaviours in complex networks (or how do individuals and collectives shape each other).

Prof. Marques-Pita was awarded a PhD in Artificial Intelligence and Cognitive Science by the University of Edinburgh (Scotland, UK) in 2007. He then pursued postdoctoral training with Prof. Luís Rocha (Indiana University, USA) and Prof. Melanie Mitchell (Portland State University, USA) from 2008 to 2014. During this period, he studied dynamics and control in complex networks, particularly in automata models of biochemical regulation and signalling in Biology. His current main focus is on studying complex communication processes in various types of social network, with a particular focus on group conversation. He is the principal investigator of a project that uses Artificial Intelligence, Data Science and Network Science to study social conversation in communities of secondary school students online. The main goal of the project is to use Artificial Intelligence to support the Portuguese Ministry of Education in designing targeted educational interventions that are informed by evidence, and not by intuitions, while ensuring the adherence to ethical principles, and full accountability. These interventions can be helpful for facing current challenges posed by the increasing divisive narratives, extremism, conspiracy theories and hate speech that are becoming so dangerously commonplace in our societies. We expect to provide such support by informing decision makers on how behaviours such as reflection, argumentation, critical thinking, and allowing a multiplicity of voices and perspectives can be successfully encouraged in a diversity of educational settings, in a constantly changing social reality.

Prof. Marques-Pita is also involved in similar projects that use large datasets to understand collective information processing dynamics in specific conversation topics in online social-network platforms, particularly on topics that are directly or indirectly related to politics.


  • Collective information processing and control in complex networks
  • Data-driven analysis of conversation in social networks
  • Ethical and accountable uses of AI to support e.g. decisions affecting people and modelling of individual behaviours
  • Conversational interaction between people and public spaces


  • PhD in Artificial Intelligence and Cognitive Science, 2007

    University of Edinburgh

Recent Publications

Quickly discover relevant content by filtering publications.

PTPARL-D: Annotated Corpus of 44 years of Portuguese Parliament debates

Here, we present PTPARL-D, an annotated corpus of debates in the Portuguese Parliament, from 1976 to 2019, covering the entire period of Portuguese democracy.

IoT and Engagement in the Ubiquitous Museum

We installed a simple location sensing network in an Italian musem and collected a sparse, non-overlapping dataset of individual visits to the twenty rooms of a curated classical art exhibition. We found that the choice of rooms visited, sequence and time spent in each of them was not random. Indeed, most visits could be described as instances of a few linearly separable patterns.

Early and Real-Time Detection of Seasonal Influenza Onset

We propose a new method that identifies the beginning of the yearly flu season. This is done by using several different data sources, including searches for flu-related symptoms on Google and phone call logs to a specialized medical phone service.

The effective structure of complex networks drives dynamics, criticality and control

We show that in general the control of complex networks cannot be predicted from structure alone. Structure-only methods such as structural controllability and minimum dominating set theory both undershoot and overshoot the number and which sets of variables actually control these models, highlighting the importance of dynamics in determining control. We show that canalization measured as logical redundancy in automata transition functions models plays a very important role in the extent to which structure predicts dynamics

The effective structure of complex networks: Canalization in the dynamics of complex networks drives dynamics, criticality and control

Network Science has provided predictive models of many complex systems from molecular biology to social interactions. Most of this success is achieved by reducing multivariate dynamics to a graph of static interactions. Such network structure approach has provided many insights …