Resumo:
Electronic Participation, or e-Participation, is defined as citizens’ participation in decisionmaking
from public administration using Information and Communications Technologies.
The e-Participation is a multidisciplinary research field, with contributions from areas
such as political science, sociology, administration, psychology, and economics, as well as
contributions of a more technical nature, such as computer science. The field of psychology,
specifically, allows the investigation of human personality and its motivations, providing
structural representations of personality traits and making it possible to describe
an individual’s intentions to participate in these environments. Thus, this work aims to
study computational techniques for identifying and analyzing personality traits in Electronic
Participation environments. To aim this, Machine Learning algorithms, specifically
regression models, and the Five-Factor Model, also known as the Big Five, were used.
Analyzing the results, it was found that the Random Forest model had the best performance,
with a mean absolute error of 0.02619. In addition, in the context of the tool
analyzed, the personality trait that stands out the most is Openness, accompanied by
Conscientiousness and Agreeableness. Extraversion and Neuroticism traits appear with
lower scores.