Submitted by Erika Demachki (erikademachki@gmail.com) on 2016-03-17T17:31:58Z No. of bitstreams: 2 Dissertação - Santiago Valdés Ravelo - 2011.pdf: 730949 bytes, checksum: 92c89c8c1f240082004834898896b9ba (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Erika Demachki (erikademachki@gmail.com) on 2016-03-17T17:35:15Z (GMT) No. of bitstreams: 2 Dissertação - Santiago Valdés Ravelo - 2011.pdf: 730949 bytes, checksum: 92c89c8c1f240082004834898896b9ba (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2016-03-17T17:35:15Z (GMT). No. of bitstreams: 2 Dissertação - Santiago Valdés Ravelo - 2011.pdf: 730949 bytes, checksum: 92c89c8c1f240082004834898896b9ba (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2011-02-18 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work considers three relevant NP-hard problems. The firstone is the one-dimensional cutting stock problem in which the non-used material in the cutting patterns may be used in the future. For this problem we analyze the existing mathematical models, propose new models, design a heuristic and two metaheuristic approaches, being their performances improved by using parallel programming, and solve instances, practical and randomly generated, from the literature. The computational experiments were quite good for all tested instances. The second problem we consider is the stable roommates problem (a variant of the stable matching problem). For this we give two mathematical programming models, sequential and parallel implementations of a Tabu Search, and a Branch-andBound. Also, we report computational experiments to instances of the problem. The last problem we consider is the compartmentalized knapsack problem (a generalization of the knapsack problem) for which we analyze a quadratic integer model and give a linear integer model. We design a greedy heuristic and a GRASP algorithm, that uses path-relinking, and solve randomly generated instances. All parallel implementations use Graphics Processing Units (GPUs). / Este trabalho considera três problemas, NP-difíceis, relevantes de estudo em otimização combinatória. O primeiro deles é o problema de corte uni-dimensional de objetos, onde o material não usado pelos padrões de corte pode ser usado no futuro. Para este problema analisamos os modelos matemáticos existentes, propomos novos modelos, projetamos uma heurística construtiva e duas metaheurísticas, sendo seus desempenhos melhorados com programação paralela, e resolvemos instâncias, práticas e aleatórias, encontradas na literatura; sendo os experimentos computacionais muito bons para todas as intânciastestadas.Osegundoproblemaqueconsideramoséoproblemadoscompanheiros estáveis (stable roommates problem), uma variante do problema de emparelhamento estável (stable matching problem). Para este propomos dois modelos matemáticos, uma implementação sequencial e uma paralela de uma Tabu Search, e um Branch-andBound. Também reportamos experimentos computacionais para instâncias do problema. O último problema considerado é o da mochila compartimentada (uma generalização do problema clássico da mochila), para o qual analisamos uma modelagem quadrática inteira e propomos um modelo linear inteiro; também projetamos uma heurística gulosa, um algoritmo GRASP, que usa path-relinking, e resolvemos intâncias geradas aleatóriamente. Todas as implementações em paralelo usam unidades de processamento gráfico (Graphics Processing Units, GPUs).

De nos jours, avec l'utilisation croissante des technologies numériques, l'éducation à la préservation de la vie privée joue un rôle important en particulier pour les adolescents. Bien que plusieurs plateformes d'apprentissage en ligne à la sensibilisation à la vie privée aient été mises en œuvre, elles sont généralement basées sur des techniques traditionnelles d'apprentissage. Plus particulièrement, ces plateformes ne permettent pas aux étudiants de coopérer et de partager leurs connaissances afin d’améliorer leur apprentissage ensemble. En d'autres termes, elles manquent d'interactions élève-élève. Des recherches récentes sur les méthodes d'apprentissage montrent que la collaboration entre élèves peut entraîner de meilleurs résultats d'apprentissage par rapport à d'autres approches. De plus, le domaine de la vie privée étant fortement lié à la vie sociale des adolescents, il est préférable de fournir un environnement d'apprentissage collaboratif où l’on peut enseigner la préservation de la vie privée, et en même temps, permettre aux étudiants de partager leurs connaissances. Il serait souhaitable que ces derniers puissent interagir les uns avec les autres, résoudre des questionnaires en collaboration et discuter de problèmes et de situations de confidentialité. À cet effet, ce travail propose « Teens-online », une plateforme d'apprentissage en ligne collaborative pour la sensibilisation à la vie privée. Le programme d'études fourni dans cette plateforme est basé sur le Référentiel de formation des élèves à la protection des données personnelles. De plus, la plateforme proposée est équipée d'un mécanisme d'appariement de partenaires basé sur la théorie des jeux. Ce mécanisme garantit un appariement élève-élève stable en fonction des besoins de l'élève (comportement et / ou connaissances). Ainsi, des avantages mutuels seront obtenus en minimisant les chances de coopérer avec des pairs incompatibles. Les résultats expérimentaux montrent que l'utilité moyenne obtenue en appliquant l'algorithme proposé est beaucoup plus élevée que celle obtenue en utilisant d'autres mécanismes d'appariement. Les résultats suggèrent qu'en adoptant l'approche proposée, chaque élève peut être jumelé avec des partenaires optimaux, qui obtiennent également en retour des résultats d'apprentissage plus élevés. / Nowadays, with the increasing use of digital technologies, especially for teenagers, privacy education plays an important role in their lives. While several e-learning platforms for privacy awareness training have been implemented, they are typically based on traditional learning techniques. In particular, these platforms do not allow students to cooperate and share knowledge with each other in order to achieve mutual benefits and improve learning outcomes. In other words, they lack student-student interaction. Recent research on learning methods shows that the collaboration among students can result in better learning outcomes compared to other learning approaches. Motivated by the above-mentioned facts, and since privacy domain is strongly linked to the social lives of teens, there is a pressing need for providing a collaborative learning platform for teaching privacy, and at the same time, allows students to share knowledge, interact with each other, solve quizzes collaboratively, and discuss privacy issues and situations. For this purpose, this work proposes “Teens-online”, a collaborative e-learning platform for privacy awareness. The curriculum provided in this platform is based on the Personal Data Protection Competency Framework for School Students. Moreover, the proposed platform is equipped with a partner-matching mechanism based on matching game theory. This mechanism guarantees a stable student-student matching according to a student's need (behavior and/or knowledge). Thus, mutual benefits will be attained by minimizing the chances of cooperating with incompatible students. Experimental results show that the average learning-related utility obtained by applying the proposed partner-matching algorithm is much higher than the average utility obtained using other matching mechanisms. The results also suggest that by adopting the proposed approach, each student can be paired with their optimal partners, which in turn helps them reach their highest learning outcomes.