Principal Investigator

Pedro Brandão, (ISTEC Lisbon);
José Vicente dos Reis, (ISTEC Lisbon);
José Almeida, MSc student (ISTEC Lisbon);
Marcelo Camerlo, ICS;
António Costa Pinto, ICS.

Start Date

January 1st 2022

Inserted in the United Nations Sustainable Development Goals (Agenda 2030):

Goal 4. Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all;
Goal 16. Promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build effective, accountable and inclusive institutions at all levels;
Goal 17. Strengthen the means of implementation and revitalize the global partnership for sustainable development.

The research project’s main goal is to produce data visualization in an intuitive way in several digital media.
The data to be processed for visualization purposes are from the South European Government Project, coordinated by Marcelo Camerlo and António Costa Pinto, in the area of elites study developed at ICS. This data has a regional coverage, Southern Europe (Portugal, Spain, Italy, Greece and Turkey), the temporal coverage is from 2000 to 2021. The dimensions covered are:

Personal background (gender, age);
Expertise (experience in positions related to the area of the portfolio held, including the public and private sector);
Political competencies (experience in political positions in the legislative and executive arenas at the national, international, and local level);
Partisanship (affiliation, seniority);
Ministerial portfolios (occupation, duration).
Data visualization through appropriate and purpose-built tools are extremely important to be able to draw quick and intuitive conclusions and still conclusions that may be hidden in the data because they are not standardized. Therefore, digital visualization of data can create patterns that allow advanced conclusions to be drawn. This is what is intended to be done in the present research work.
Data visualizations are surprisingly common in everyday life, but usually appear in the form of familiar tables and graphs. A combination of multiple visualizations and bits of information is often referred to as infographics.
Data visualizations can be used to discover unknown facts and trends. We can see visualizations in the form of line graphs to display changes over time. Bar and column charts are useful when looking at relationships and making comparisons. Pie charts are a great way to show parts-of-a-thing. But so are data infographics, which are the best way to visually share data of any kind.
Good data visualizations are created when communication, data science, and design collide. Data visualizations done right offer important information about complicated data sets in meaningful and intuitive ways. American researcher and Yale professor Edward Tufte argues that excellent data visualizations consist of “complex ideas communicated with clarity, precision, and efficiency,” Image 1.

Being able to understand and read data visualizations has become a necessary requirement for the 21st century. As data visualization tools and resources have become readily available, more and more professionals are expected to be able to collect and visually treat information from data.

Principal Investigators

Pedro Brandão
João Monge

Collaborators

Master in Computer Science Students
Starting Date

1st of July, 2021

It is intended to develop an e-mail client with an Anti-Spam and anti-Phishing module, which can be installed and used by ISTEC staff. An unprecedented Machine Learning algorithm will be developed.
In today’s world, e-mail plays a fundamental role, regardless of the area of work. In this scenario, phishing emails are one of the main threats in today’s world. These emails “look” legitimate, but lead users to malicious websites. As a result, the user, organization or institution ends up being the object of online hackers. To tackle these problems, various statistical methods have been applied and developed. In this paper, we will make use of the distributional representation, namely TF-IDF, for the numerical representation of phishing emails. A comparative study of classical machine learning techniques, such as Random Forest, AdaBoost, Naive Bayes, Decision Tree, SVM, will also be used.

Principal Investigator

João Carneiro

Researchers

Pedro Brandão
Diogo Algarvio

Collaborators

MSc in Informatics Students

Starting Date

19th of July, 2021

The human brain, with about 86 billion neurons and 7000 connections, is an extraordinary complex system with a high degree of complexity. One of the higher functions that the brain enables is vision. This project, aims to consider how to encode the visual signal from the biological retina to inspire robotic applications that have high speed requirements.
Neuromorphic Engineering is the domain that draws on known models from Neuroscience to develop problems in Engineering and Robotics. Neuromorphic retinas inspired by the human retina encode visual information differently from conventional cameras.
The project aims to evaluate whether the use of bio-inspired models allows for modeling escape reflexes by detecting objects moving toward the robot at high speed. Existing challenges include the motion of the robot or whether the accuracy of the cameras will be sufficient. Studies with insects show that they have visual systems that are quite effective at modeling these reflexes. However, the question of whether these behaviors can be adapted for robots still remains.

Researchers

Pedro Brandão
Sérgio Pinto

Starting Date

September 2022

The goal is to develop an application to analyze/audit potential information security and cybersecurity risks for a given organization and thus assist the organization in choosing the security measures and controls to define and implement to ensure an adequate level of security.
To this end, the application should support an interactive questionnaire on configurations used in components/assets to be analyzed (accounting for the respective controls, threats and current vulnerabilities), according to security reference recommendations.
For these recommendations the reference document “Guide for Risk Management” of the CNSC should be followed, which incorporates guidelines from the QNRCS and RJSC. This document refers to the need to establish the context of the risk analysis, followed by the following steps:

1) Risk identification,

2) Risk Analysis, and

3) Risk assessment.

This last step should lead to a conclusion with the recommendations of necessary actions (or not) to be taken to ensure a level of safety appropriate to the risks in question.