Cookies
O website necessita de alguns cookies e outros recursos semelhantes para funcionar. Caso o permita, o INESC TEC irá utilizar cookies para recolher dados sobre as suas visitas, contribuindo, assim, para estatísticas agregadas que permitem melhorar o nosso serviço. Ver mais
Aceitar Rejeitar
  • Menu
Sobre

Sobre

Dennis recebeu o grau de Mestre em Engenharia Informática na UTAD em Dezembro de 2018. Ele participou desde Setembro de 2016 até Dezembro de 2018 no projeto NanoStima RL2 - Passus Mobile, sendo responsável pelo desenvolvimento de um sistema que supervisiona o exercício nas pessoas com doença arterial periférica. No período entre Janeiro de 2019 e Novembro de 2019, fez parte do projeto Ecsaap, com o objetivo de desenvolver um sistema informático para a visualização e deteção de fenómenos metereológicos.

Em Dezembro de 2019 começou o Doutoramento em Informática na UTAD, com bolsa financiada pela FCT.

Tópicos
de interesse
Detalhes

Detalhes

  • Nome

    Dennis Lourenço Paulino
  • Cluster

    Informática
  • Cargo

    Assistente de Investigação
  • Desde

    01 setembro 2016
001
Publicações

2023

A Model for Cognitive Personalization of Microtask Design

Autores
Paulino, D; Guimaraes, D; Correia, A; Ribeiro, J; Barroso, J; Paredes, H;

Publicação
SENSORS

Abstract
The study of data quality in crowdsourcing campaigns is currently a prominent research topic, given the diverse range of participants involved. A potential solution to enhancing data quality processes in crowdsourcing is cognitive personalization, which involves appropriately adapting or assigning tasks based on a crowd worker's cognitive profile. There are two common methods for assessing a crowd worker's cognitive profile: administering online cognitive tests, and inferring behavior from task fingerprinting based on user interaction log events. This article presents the findings of a study that investigated the complementarity of both approaches in a microtask scenario, focusing on personalizing task design. The study involved 134 unique crowd workers recruited from a crowdsourcing marketplace. The main objective was to examine how the administration of cognitive ability tests can be used to allocate crowd workers to microtasks with varying levels of difficulty, including the development of a deep learning model. Another goal was to investigate if task fingerprinting can be used to allocate crowd workers to different microtasks in a personalized manner. The results indicated that both objectives were accomplished, validating the usage of cognitive tests and task fingerprinting as effective mechanisms for microtask personalization, including the development of a deep learning model with 95% accuracy in predicting the accuracy of the microtasks. While we achieved an accuracy of 95%, it is important to note that the small dataset size may have limited the model's performance.

2023

Investigating Author Research Relatedness through Crowdsourcing: A Replication Study on MTurk

Autores
Correia, A; Paulino, D; Paredes, H; Guimarães, D; Schneider, D; Fonseca, B;

Publicação
26th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2023, Rio de Janeiro, Brazil, May 24-26, 2023

Abstract

2023

Cognitive personalization for online microtask labor platforms: A systematic literature review

Autores
Paulino, D; Correia, A; Barroso, J; Paredes, H;

Publicação
USER MODELING AND USER-ADAPTED INTERACTION

Abstract
Online microtask labor has increased its role in the last few years and has provided the possibility of people who were usually excluded from the labor market to work anytime and without geographical barriers. While this brings new opportunities for people to work remotely, it can also pose challenges regarding the difficulty of assigning tasks to workers according to their abilities. To this end, cognitive personalization can be used to assess the cognitive profile of each worker and subsequently match those workers to the most appropriate type of work that is available on the digital labor market. In this regard, we believe that the time is ripe for a review of the current state of research on cognitive personalization for digital labor. The present study was conducted by following the recommended guidelines for the software engineering domain through a systematic literature review that led to the analysis of 20 primary studies published from 2010 to 2020. The results report the application of several cognition theories derived from the field of psychology, which in turn revealed an apparent presence of studies indicating accurate levels of cognitive personalization in digital labor in addition to a potential increase in the worker's performance, most frequently investigated in crowdsourcing settings. In view of this, the present essay seeks to contribute to the identification of several gaps and opportunities for future research in order to enhance the personalization of online labor, which has the potential of increasing both worker motivation and the quality of digital work.

2023

Exploring Stigmergic Collaboration and Task Modularity Through an Expert Crowdsourcing Annotation System: The Case of Storm Phenomena in the Euro-Atlantic Region

Autores
Paulino, D; Correia, A; Yagui, MMM; Barroso, J; Liberato, MLR; Vivacqua, AS; Grover, A; Bigham, JP; Paredes, H;

Publicação
IEEE ACCESS

Abstract
Extreme weather events, such as windstorms, hurricanes, and heat waves, exert a significant impact on global natural catastrophes and pose substantial challenges for weather forecasting systems. To enhance the accuracy and preparedness for extreme weather events, this study explores the potential of using expert crowdsourcing in storm forecasting research through the application of stigmergic collaboration. We present the development and implementation of an expert Crowdsourcing for Semantic Annotation of Atmospheric Phenomena (eCSAAP) system, designed to leverage the collective knowledge and experience of meteorological experts. Through a participatory co-creation process, we iteratively developed a web-based annotation tool capable of capturing multi-faceted insights from weather data and generating visualizations for expert crowdsourcing campaigns. In this context, this article investigates the intrinsic coordination among experts engaged in crowdsourcing tasks focused on the semantic annotation of extreme weather events. The study brings insights about the behavior of expert crowds by considering the cognitive biases and highlighting the impact of existing annotations on the quality of data gathered from the crowd and the collective knowledge generated. The insights regarding the crowdsourcing dynamics, particularly stigmergy, offer a promising starting point for utilizing stigmergic collaboration as an effective coordination mechanism for weather experts in crowdsourcing platforms but also in other domains requiring expertise-driven collective intelligence.

2022

Uncovering the Potential of Cognitive Personalization for UI Adaptation in Crowd Work

Autores
Paulino, D; Correia, A; Guimarães, D; Barroso, J; Paredes, H;

Publicação
25th IEEE International Conference on Computer Supported Cooperative Work in Design, CSCWD 2022, Hangzhou, China, May 4-6, 2022

Abstract
Crowdsourcing has received considerable attention over the last fifteen years and has been the subject of several experiments that demonstrate its large potential for use in real-world situations. With the rapid growth of and access to crowd work environments, there is a need for new ways to ensure more equitable access for all people. Task design is one of the core aspects of the crowdsourcing process and its optimization is a priority for many requesters that want to have their tasks solved in short times and with high levels of accuracy. Aligned with this goal, a cognitive personalization framework can make it feasible to assess the information processing preferences of crowd workers in order to provide a useful user interface (UI) adaptation. In an effort to address this issue, this study recruited a total of 64 crowd workers to take cognitive style tests and perform prototypical tasks. The results indicate that it is possible to apply short tests and then obtain some useful indicators for better matching tasks to workers with implications for improving the general outcomes and acceptance rates in crowdsourcing.