[pt] Big Data é o termo utilizado para caracterizar o conjunto de soluções tecnológicas que permitem o rápido processamento de um grande volume de dados variados. Estas soluções só se tornaram possíveis com os avanços tecnológicos ocorridos nas últimas décadas. Uma das funcionalidades que ganharam força e melhorias através desses tipos de tecnologias são os sistemas de recomendação. Tais sistemas têm como objetivo principal oferecer ao usuário sugestões de conteúdo que possam interessá-lo. Este conteúdo pode ser uma notícia, um produto, um contato, um filme, uma música ou qualquer outro tipo de informação. Esta dissertação estuda a percepção dos usuários em relação aos sistemas de recomendação, especialmente para o conteúdo televisivo (programas, séries e filmes). Para tal, fez-se uso de questionários, grupos de foco, análise do cenário atual e estudo de caso. Através destes métodos e técnicas foi possível identificar os diferentes fatores que influenciam a maneira como a funcionalidade é percebida e a forma como os serviços são utilizados. Além disso, se discute as consequências do uso excessivo da personalização de conteúdo, bem como questões éticas, privacidade, impactos sociais e psicológicos e a responsabilidade do designer de produtos digitais. Em conclusão, são feitas recomendações para o desenvolvimento deste tipo de sistema de forma que atenda aos seus objetivos e proporcione uma experiência mais satisfatória ao usuário. / [en] Big Data is the term used to identify the set of technological solutions that allows the fast processing of a big amount of diverse data that only became possible with the technological advances that have occurred in recent decades. One of the features that gained strength and improvements through these types of technologies are the recommendation systems. The objective of these kind of systems is to offer suggestions of content that might interest the users. This content can be some news, a product, a personal contact, a movie, a song or any other kind of information. This dissertation addresses the study of the perception of the users relative to recommendation systems, especially for television content (programs, series and movies). For this purpose, questionnaires, focus groups, context analysis and case studies were used. Through these methods and techniques it was possible to identify the different factors that influences how the functionality is perceived and how the services are used. Further, it discusses the consequences of the excessive use of personalized content, privacy, ethical and social issues, psychological impacts and the responsibility of the digital products designer. In conclusion, some recommendations are made regarding the development of this type of system so that it achieves its purposes and provides a more satisfying user experience.

[pt] Hoje em dia e-learning está se tornando mais importante por possibilitar a disseminação de conhecimento e informação através da internet de uma forma mais rápida e menos dispendiosa. Consequentemente, de modo a filtrar o que é mais relevante e/ou de interesse do usuário, arquiteturas e técnicas de personalização vêm sendo abordadas. Dentre as muitas possibilidades de personalização existentes, a que lida com acessibilidade está se tornando essencial, pois garante que uma grande variedade de usuários possa ter acesso à informação conforme suas necessidades e características. Acessibilidade não é apenas garantir que pessoas com alguma deficiência, ou dificuldade, possam ter acesso à informação, apesar de ser importante e eventualmente ser uma exigência legal. Acessibilidade é também garantir que uma larga variedade de usuários e interfaces possam obter acesso à informação, maximizando assim a audiência potencial. Esta dissertação apresenta uma extensão do LORIS, um sistema de integração de repositórios de objetos de aprendizagem, descrevendo as alterações na sua arquitetura para ser capaz de lidar com acessibilidade e reconhecer diferentes versões de um mesmo objeto de aprendizagem, permitindo assim que um usuário execute uma consulta considerando seu perfil e preferências. Foi desenvolvido um protótipo dos serviços descritos na arquitetura utilizando serviços Web e navegação facetada, bem como padrões web, de e-learning e de acessibilidade. O uso de serviços Web e de padrões visa promover flexibilidade e interoperabilidade, enquanto a navegação facetada, como implementada, permite que o usuário aplique múltiplos filtros aos resultados da consulta sem a necessidade de re-submetê-la. / [en] Nowadays e-learning is becoming more important as it makes possible the dissemination of knowledge and information through the internet in a faster and costless way. Consequently, in order to filter what is more relevant and/or of users interest, architectures and personalization techniques have been raised. Among the many existing possibilities of personalization, the one that deals with accessibility is becoming essential because it guarantees that a wide variety of users may have access to the information according to their preferences and needs. Accessibility is not just about ensuring that disabled people can access information, although this is important and may be a legal requirement. It is also about ensuring that the wide variety of users and devices can all gain access to information, thereby maximizing the potential audience. This dissertation presents an extension of LORIS, an integration system of learning object repositories, describing the changes on its architecture to make it able to deal with accessibility and to recognize different versions of the same learning object, thus allowing a user to execute a query considering his/her preferences and needs. A prototype of the services that are described in the architecture was developed using web services and faceted navigation, as well as e-learning and accessibility standards. The use of web services and standards aims at providing flexibility and interoperability, while the faceted navigation, as implemented, allows the user to apply multiple filters to the query results without the need to resubmit it.

La place primordiale accordée à l’apparence physique dans la société actuelle, qui est une société de «l’image», a fait de la chirurgie esthétique la pratique chirurgicale la plus répandue et réclamée. Cette pratique, d’abord considérée illicite, puis exclusivement liée à des fins thérapeutiques, fut reconnue par la jurisprudence française à la suite d’une évolution prétorienne en 1936. Les raisons d’un tel retard de reconnaissance résident dans la spécificité qui distingue la chirurgie esthétique des autres branches de la médecine chirurgicale : en effet, la chirurgie esthétique se caractérise d’une part par sa nature qui est dépourvue de toute finalité curative et qui est réalisée sur une personne jouissant d’une bonne santé et, d’autre part, par sa finalité puisqu’elle vise à réaliser une amélioration de l’apparence physique et de la morphologie de la personne concernée. Cette double spécificité distingue également la chirurgie esthétique de la chirurgie reconstructrice et réparatrice. Ces caractéristiques mêmes de la chirurgie esthétique ont exigé, de la jurisprudence, mais également du législateur français, un traitement juridique spécial du chirurgien esthétique en vue d’éviter que ce dernier ne profite de la fragilité, aussi bien physique que psychique, de la personne ayant une obsession de beauté et de perfectionnement physique dans un domaine non curatif. C’est ainsi que nous avons abordé, dans la présente étude, les points spécifiques du traitement juridique du chirurgien esthétique, tant au regard de ses obligations que de sa responsabilité. Relativement aux obligations, la spécificité de la chirurgie esthétique exige tout d’abord un devoir d’information rigoureuse de la part du praticien afin d’aboutir à un consentement préalable éclairé du patient, ce qui suppose un devis écrit contenant tous les renseignements relatifs à l’intervention. Elle exige, d’autre part, un processus préalable en trois étapes […] Pour ce qui est de la responsabilité du chirurgien esthétique, elle est également spécifique, et ce sur les plans civil et pénal : Sur le plan civil, sa spécificité a poussé la jurisprudence française, qui a réalisé son approche du point de vue de la nature de l’obligation du chirurgien esthétique, à procéder à une démarche de « mutation » des règles applicables en droit commun dans une tentative de les adapter à la spécificité de l’intervention chirurgicale esthétique et de la responsabilité du praticien qui en découle. Cependant, la démarche jurisprudentielle était inadaptée à la spécificité de la chirurgie esthétique, et a entrainé une large confusion au sein de la jurisprudence et de la doctrine ; elle est donc critiquable. C’est ainsi que la présente étude propose un régime spécifique à la responsabilité civile du chirurgien esthétique à trois facettes : La détermination du cadre juridique de la responsabilité contractuelle du chirurgien esthétique liée à la spécificité du contrat de chirurgie esthétique ayant pour objet unique l’amélioration de morphologie et pour objectif la réalisation du résultat attendu qui a justifié l’intervention esthétique ; ainsi le chirurgien garantit le résultat esthétique de l’intervention. Le maintien de la responsabilité extracontractuelle en ce qui concerne les dommages qui surviennent à l’occasion de l’exécution de l’obligation contractuelle. L’institution d’un régime spécifique prenant en considération l’éventualité des dommages survenus en raison des aléas thérapeutiques et des risques exceptionnels, en envisageant un régime d’indemnisation automatique lié à des cas spécifiques qui se place en dehors du régime de la responsabilité civile. Quant à la responsabilité pénale du chirurgien esthétique, une incrimination spéciale et strictement adaptée à la nature de la chirurgie esthétique s’applique essentiellement dans trois circonstances : le remodelage sexuel, la publicité mensongère et l’expérimentation scientifique. / The prominent place given to physical appearance in the current society qualified as the society of the “image”, has made cosmetic surgery, the most common surgical practice and claimed. This practice which was deemed illegal at one time and was then linked to the only therapeutic purposes has been recognized by the French courts following a change praetorian in 1936. The reasons supporting such a delay of recognition result from the specificity that distinguishes cosmetic surgery of the other surgical branches, it’s characterized by its very nature which is devoid of any curative purposes and is performed on a healthy person and its purpose is to achieve an improvement in the physical appearance and the morphology of the person concerned. This dual specificity also distinguishes cosmetic surgery from reconstructive and restorative surgeries. The specifications of cosmetic surgery required a special legal treatment of the cosmetic surgeon so it does not take advantage of the physical and mental fragility that has an obsession with beauty and physical development. Thus, we approached in this study the specific features of the legal treatment of the cosmetic surgeon in terms of obligations and responsibility.Regarding obligations, the specificity of cosmetic surgery requires a rigorous information obligation of the cosmetic surgeon to lead to a prior informed consent of the patient contained also on a written quotation. It requires a three-step process prior […] As for the cosmetic surgeon’s responsibility, it is also specific regarding its civil and criminal responsibilities : Specificity of its liability has led the French courts which made its approach towards the nature of the cosmetic surgeon’s obligation to conduct a process of «changing» the rules of common law in an attempt to adapt to specificity of cosmetic surgery and the resulting liability. However, the judicial process was not adapted to the specificity of cosmetic surgery, it led to a significantly wider confusion in the jurisprudence and doctrine, it is therefore questionable. Thus, we propose in this paper a specific regime for civil liability cosmetic surgeon three-sided : The contractual liability of the cosmetic surgeon linked to the specific contract cosmetic surgery that aims to improve morphology and has the objective of achieving the expected result which has justified the intervention. Thus, the cosmetic surgeon provides the aesthetic result of the intervention. Maintaining liability regarding damages that occur during the execution of the contractual obligation. The establishment of a special regime taking into account the case of damage due to therapeutic hazards and exceptional risks considering a compensation scheme automatically linked to specific cases that are placed outside the regime of civil liability. As to the criminal responsibility of the cosmetic surgeon, a special incrimination is strictly adapted to the nature of the surgery mainly applies in three situations : sexual remodeling, false advertising and scientific experimentation.

SILVA, Fernando Wisse Oliveira. Personalização da representação política: um estudo sobre as estratégias de comunicação dos deputados Marco Feliciano e Jean Wyllys no twitter. 2015. 283f. – Dissertação (Mestrado) – Universidade Federal do Ceará, Programa de Pós-graduação em Comunicação Social, Fortaleza (CE), 2015. / Submitted by Márcia Araújo (marcia_m_bezerra@yahoo.com.br) on 2015-12-22T11:42:49Z No. of bitstreams: 1 2015_dis_fwosilva.pdf: 4031639 bytes, checksum: 16ca271872213344ad9d971d469b94eb (MD5) / Approved for entry into archive by Márcia Araújo(marcia_m_bezerra@yahoo.com.br) on 2015-12-22T12:24:49Z (GMT) No. of bitstreams: 1 2015_dis_fwosilva.pdf: 4031639 bytes, checksum: 16ca271872213344ad9d971d469b94eb (MD5) / Made available in DSpace on 2015-12-22T12:24:49Z (GMT). No. of bitstreams: 1 2015_dis_fwosilva.pdf: 4031639 bytes, checksum: 16ca271872213344ad9d971d469b94eb (MD5) Previous issue date: 2015 / O presente trabalho tem por objetivo discutir a personalização da política por parte dos representantes ao longo do mandato. Analisa-se como a utilização de media digitais tem participação nesse fenômeno. Explora-se como agentes do campo acabam se apropriando dessas redes digitais para aumentar sua visibilidade, através da promoção da própria imagem, a fim de influenciar favoravelmente a opinião pública. Esses meios são frequentemente encarados com o potencial de aproximar a política dos cidadãos, aumentando o engajamento da esfera civil. Examinando os perfis pessoais de dois deputados federais no Twitter – Marco Feliciano (PSC-SP) e Jean Wyllys (PSOL-RJ) –, busca-se estudar o uso ―personalizado‖ dos media digitais por estes agentes políticos. Para isso, foram coletados e estudados os tweets publicados nos meses de setembro, outubro e novembro de 2013, a fim de contemplar os perfis dos dois parlamentares de maneira mais contínua, sem se restringir a um acontecimento específico, enfatizando-se os perfis dos diálogos travados entre os parlamentares e os usuários. Para classificação dos tweets, utilizou-se análise de conteúdo das mensagens na tentativa de dimensionar a gestão de imagem dos deputados a partir do que foi publicado em seus perfis. A análise das mensagens revelou uma composição de uma imagem pública de dois deputados bastante integrados com as novas tecnologias e que, pelo menos na aparência, estão dispostos a manter um diálogo direto com os cidadãos. Ao gerenciarem suas imagens públicas, têm-se principalmente os deputados falando sobre as posições detidas e realizações pessoais, ligando, por vezes, a várias remodelações e compromissos que ocorrem no seu cotidiano. Portanto, trata-se de uma informação pessoal, mas de natureza profissional e não oferece a seus seguidores um senso do processo político interno para a tomada de certas decisões. Porém, é inegável que a utilização de ferramentas digitais aproxima representantes e representados e, por consequência, também aproxima o discurso político da sociedade, ainda que as trocas sejam limitadas.

Submitted by Marcia Bacha (marcia.bacha@fgv.br) on 2011-05-17T18:24:30Z No. of bitstreams: 1 000361843.pdf: 3031530 bytes, checksum: a2ebaa52251295b69fe71883cc6526e1 (MD5) / Approved for entry into archive by Marcia Bacha(marcia.bacha@fgv.br) on 2011-05-17T18:24:37Z (GMT) No. of bitstreams: 1 000361843.pdf: 3031530 bytes, checksum: a2ebaa52251295b69fe71883cc6526e1 (MD5) / Approved for entry into archive by Marcia Bacha(marcia.bacha@fgv.br) on 2011-05-17T18:24:56Z (GMT) No. of bitstreams: 1 000361843.pdf: 3031530 bytes, checksum: a2ebaa52251295b69fe71883cc6526e1 (MD5) / Made available in DSpace on 2011-05-17T18:25:05Z (GMT). No. of bitstreams: 1 000361843.pdf: 3031530 bytes, checksum: a2ebaa52251295b69fe71883cc6526e1 (MD5) Previous issue date: 2004 / This dissertation aims at evaluating how the enterprises organize their strategies for the management of knowledge. Basically there are two strategic models presented in the literature: codification and personalization. However, there is an issue that is presented in the literature regarding the importance of prioritizing focus in only one of the mentioned strategic models. The main objective is to present a rank that is meant to analyze whether a leader enterprise in its segment has a strategic focus regarding the management ofknowledge. During the development of this work a theoretical reference will be followed in search of an explanation for the importance of the management of knowledge in the enterprises strategy. Points such as the characteristics of the Information Era and the basic concepts of the management of knowledge will be studied, comprising theoretical models that seek to explain how the knowledge is created and transferred within the enterprises. A model for measuring the strategic focus of an enterprise will be proposed. This model will be composed of structured indicators in form of a questionnaire. For every indicator there will be an evaluation scale and the questionnaire set will search the translation of the mentioned strategic focus. To sum up, the questionnaire will be tested in an Intellectual Property enterprise and statistic analyzes to evaluate it will be made. To achieve this goal the confirmatory factorial analysis procedure will be used. This model is a statistic tool that makes the evaluation of the analysis model as well as the structural model possible. / Esta dissertação objetiva avaliar como as empresas organizam sua estratégia para gestão do conhecimento. Basicamente dois modelos estratégicos são apresentados na literatura: a codificação e a personalização. No entanto, há uma questão que se apresenta na literatura que se refere à importância do foco prioritário em apenas um dos modelos estratégicos citados. Como objetivo principal, pretende-se apresentar uma escala que possa analisar se uma empresa líder em seu segmento realmente possui um foco estratégico com relação à gestão do conhecimento. Durante o desenvolvimento deste trabalho, será percorrido um referencial teórico que busca explicar como a gestão do conhecimento se tomou importante para a estratégia das empresas, tocando em pontos como características da Era da Informação e conceitos básicos da gestão do conhecimento, englobando modelos teóricos que buscam explicar como o conhecimento é criado e transferido dentro das empresas. Um modelo para mensuração do foco estratégico de uma empresa será proposto. Este modelo será composto de indicadores estruturados em forma de um questionário. Para cada indicador haverá uma escala de avaliação e o conjunto do questionário buscará traduzir o foco estratégico citado. Por fim, o questionário será testado em uma empresa de Propriedade Intelectual e serão feitas análises estatísticas para avaliação do mesmo. Para isto será utilizado o procedimento da análise fatorial do tipo confirmatória. Trata-se de uma ferramenta estatística que possibilita uma avaliação tanto do modelo de análise, quanto do modelo estrutural.

Cognitive behaviour therapy and interpersonal psychotherapy are the leading treatments for eating disorders. Little is understood regarding their mechanisms of action. The research described in this thesis investigated the purported mechanisms of action of two transdiagnostic versions of these treatments; enhanced cognitive behaviour therapy (CBT-E) and interpersonal psychotherapy for eating disorders (IPT-ED). A series of mediation studies were embedded within a randomised controlled trial comparing CBT-E and IPT-ED in a transdiagnostic sample of patients with eating disorders. An analytic strategy using multilevel and structural equation modelling was used to assess for statistical mediation. Three of the key purported mediators of action of CBT-E (regularity of eating, frequency of weighing, and frequency of shape checking) and the key purported mediator of IPT-ED (interpersonal problem severity) were studied. Only regularity of eating demonstrated a strong case for being a mediator of the effect of CBT-E (on frequency of binge eating). The findings were inconclusive regarding other putative mediators of the effect of CBT-E, and for interpersonal problem severity being a mediator of the effect of IPT-ED. Limitations of the research included the non-optimal choice of measures and non-optimal timing of measurements for establishing temporal precedence. Future research should investigate the mediating role of both cognitive (e.g. interpretation of weight) and behavioural processes (e.g. frequency of weighing) in CBT-E, and the role of interpersonal functioning in CBT-E and IPT-ED. Research should use daily, or session by session measurement to better assess the temporal relationship between the purported mediator/s and the outcome/s. Experimental designs comparing dismantled versions of treatment would also help determine the relative effects that different treatment procedures have on treatment outcome/s.

Les systèmes multi-agents (SMA) mettent en relations des entités autonomes, les agents, dont les actes et les interactions participent à la mise en place d'un comportement et d'une fonctionnalité globale du système. La structure et le fonctionnement global du système sont décrits à travers la notion d'organisation : celle-ci regroupe l'ensemble des places occupées par les agents, de leurs relations et de leur importance dans la fonctionnalité attendue du système. Usuellement, ces concepts sont formalisés par des notions empruntées à l'analyse des organisations humaines et animales : rôles, groupes, normes, etc. Cependant, une part importante des organisations des SMA n'est que partiellement décrite par ces notions : la dynamique. En effet, dans les SMA ouverts, dans lesquels des agents, parfois inconnus à la conception du système, peuvent entrer ou sortir de celui-ci à tout instant, et dans les SMA auto-organisés, où la structure du système se forme et se reforme en fonction du contexte, la dynamique des organisations, c'est-à-dire l'analyse de leur évolution au cours du temps, est un champ important qui est difficile à décrire avec les notions de haut-niveau usuellement utilisées pour formaliser les organisations. Ce que nous proposons dans cette thèse est donc un modèle de description des organisations pour les SMA, axé sur la possibilité de décrire une vaste variété de système, à la fois dans leurs aspects statiques et dynamiques. Pour atteindre cela, nous nous appuyons sur une conception émergentiste et calculatoire, et nous créons notre modèle autour de trois éléments statiques typés et de bas-niveau : les agents, les relations et les tâches, et un élément dynamique : les mécanismes organisationnels. Nous proposons ensuite des méthodes descriptives des organisations fondée sur l'agrégation de contraintes sur les systèmes exprimées à partir de ces éléments, ainsi que des méthodes calculatoires destinées à permettre l'extraction de propriétés globales sur les organisations à partir de ces descriptions ; tout cela dans l'optique de fournir une aide méthodologique à la conception et à l'analyse d'organisations pour les SMA. Nous appliquons ensuite l'utilisation de notre modèle sur des applications précises afin de démontrer son intérêt et sa pratique dans la formalisation et dans la comparaison d'organisations dynamiques pour les SMA. / Multiagent systems (MAS) are composed of interacting autonomous entities called agents. Their behaviors and interactions take part in the generation of a global functionality in the system. The notion of organization describes the structural and functional aspects of such systems: it includes the objectives of each agent, the way they can interact and create relationships and their importance in the system's global function. These concepts are usually formalized through notions derived from human and animal organizations: roles, groups, norms… However, an important part of MAS' organization can only be partially described with these notions: dynamics. In open MAS – where agents can enter or exit the system at any time, and where their number and characteristics are often not known at the time of the system's design – and in self-organized MAS – where the system's structure evolves with relation to context and environment – organization's dynamics is very difficult to formally describe with these high-level notions. In this thesis, we propose a model for MAS organizations' description, which is focused toward the description of a wide range of systems and the modeling of both their static and dynamic aspects. In order to achieve this, we ground our model on an approach based on emergence and computation. We then create a model based on three basic, low-level, typed static elements: agents, relations and tasks, and one low-level dynamic element: organizational mechanisms. We then propose some methods for organization description based on our model and the principle of system's constraints aggregation. We also provide some computational methods allowing the calculation of some global properties over described organizations, in order to provide a methodological help for MAS design and analysis. We then apply our model and the proposed methods on practical applications in order to show its pertinence in MAS organizations' formalization and comparison.

Ces travaux traitent du suivi de l'activité humaine à travers l'analyse en temps r éel de signaux physiologiques et d'accélé rométrie. Il s'agit de données issues de capteurs ambulatoires ; elles sont bruitées, ambigües, et ne représentent qu'une vision incomplète de la situation. De par la nature des données d'une part, et les besoins fonctionnels de l'application d'autre part, nous considérons que le monde des possibles n'est ni exhaustif ni exclusif, ce qui contraint le mode de raisonnement. Ainsi, nous proposons un raisonnement abductif à base de modèles interconnectés et personnalisés. Ce raisonnement consiste à manipuler un faisceau d'hypothèses au sein d'un cadre dynamique de contraintes, venues tant de l'observateur (en termes d'activités acceptables) que d'exigences non-fonctionnelles, ou portant sur la santé du sujet observé. Le nombre d'hypothèses étudiées à chaque instant est amené à varier, par des mécanismes de Pr édiction-Vérification ; l'adaptation du Cadre participe également à la mise en place d'un pilotage sensible au contexte. Nous proposons un système multi-agent pour représenter ces hypothèses; les agents sont organisés autour d'un environnement partagé qui leur permet d' échanger l'information. Ces échanges et, de manière générale, la détection des contextes d'activation des agents, sont régis par des filtres qui associent une action à des conditions. Le mode de raisonnement et l'organisation de ces agents hétérogènes au sein d'un cadre homogène confèrent au système expressivité, évolutivité et maîtrise des coûts calculatoires. Une implémentation utilisant des données réelles permet d'illustrer les qualités de la proposition. / This proposal deals with human activity monitoring, through the real-time analysis of both physiology data and accelerometry. These data come from ambulatory sensors ; they are noisy and ambiguous, and merely represent a partial and incomplete observation of the current si- tuation. Given the nature of the data on one hand, and the application's required features on the other hand, we consider an Open World of non-exclusive possible situations. This has a restrictive impact on the reasoning engine. We thus propose to use abductive reasoning, based on interconnected and personalized models. This way of reasoning consists in handling a beam of hypotheses, within a dynamic Frame of constraints which come both from the Observer (who defines acceptable situations) and from non-functional expectations, or relating to the observed person's health. The number of hy- potheses at each timestep is wont to vary, by means of Prediction-Verification schemes. The evolution of the Frame leads to context-sensitive adaptive control. We propose a multi-agent system to manage these hypotheses; the agents are organized around a shared environment which allows them to trade information. This interaction and the general detection of activation contexts for the agents are powered and regulated by condition- action filters. The way of reasoning and the organization of heterogeneous agents within a homogeneous Frame lead to a system which we claim to be expressive, evolutive and cost-efficient. An imple- mentation using real sensor data is presented to illustrate these qualities.

The thesis is in the field of information design and specializing on spatial design. The study explores how workplace personalization can take form in an activity-based workplace with flexi-desks. At the time of writing, Mälardalen University is planning to build a new campus in Eskilstuna. The new workplace for the teachers and researchers is planned to be an activity-based workplace with flexi-desks. Earlier studies have shown that in many cases activity-based workplaces have not been used in the way they were designed for. In those cases, there appears to be issues with employees interaction with the physical environment.   The aim of the study has been to investigate what impact and need of workplace personalization have to the teachers in the academy of innovation, design and technology at Eskilstuna University, in order to avoid above named problems with the teachers new activity-based workplace.   Early empirical data in this study is based on interviews, to get the overall picture and to understand how activity-based workplaces work in practice at other organizations. The interviews also defined the functions of the new campus in Eskilstuna, that lead to a mapping of activity zones in the blueprints of the new building. Through studies of literature, a method has been developed that identifies the underlying needs of workplace personalization for the users in this study. The empirical data, together with earlier studies and design theory that puts human behavior first, resulted in a design proposal that satisfies the users needs for storage and personalization. The design proposal is a complement to the flexible workplace.

Made available in DSpace on 2016-06-02T19:03:59Z (GMT). No. of bitstreams: 1 6340.pdf: 3337556 bytes, checksum: 693a6a9cfb4dc2a26651724099fcf890 (MD5) Previous issue date: 2014-02-28 / Financiadora de Estudos e Projetos / Recommender Systems have a very well defined function: recommend something to someone. Through Artificial Intelligence techniques, more particularly from areas such as Data Mining and Machine Learning, it is possible to build recommendation systems. These systems will analyze large amounts of data and will inform users about some items that will probably interest them. However, some limitations of the recommender systems, which are sometimes, caused by the Mining or Learning models themselves or by the lack of available data make them computationally expensive or inaccurate. Besides, recommender systems in real environments are dynamic: data change over time or with new ratings, new users, new items or when user updates previous ratings. The Never Ending-Learning Approach (NEL) aims at a self-supervised and self-reflexive learning to mainly maximize learning of a system based on data from several sources, algorithms that can cooperate to make a better knowledge base considering the dynamic of real learning problems: learning improves along the time. As mentioned before, recommender systems are dynamic and depend on data between user and items. In order to minimize this dependency and to provide meaningful and useful results to users, this work presents a Recommender System approach guided by NEL Principles. Results show that it is possible to minimize or delay the data dependency through classifiers coupling techniques and concept deviation control. Due to that, it is possible to start with little data from a recommender system that will be dynamic and will receive new information. These new information will help even more in controlling the concept deviation and promoting the most useful recommendations. Then, this thesis presents how the Recommender System guided by NEL principles can contribute to the state of the art in recommender systems and implement a system with practical results through the Never-Ending Learning Approach. / Os Sistemas de Recomendação possuem uma função muito bem definida: recomendar algo a alguém. Através de técnicas de Inteligência Artificial, mais particularmente de áreas como a Mineração de Dados e o Aprendizado de Máquina é possível construir Sistemas de Recomendação que analisem grandes volumes de dados e consigam predizer aos usuários algo que provavelmente irá lhes interessar. No entanto, algumas limitações dos Sistemas de Recomendações, causadas as vezes pelos próprios modelos de Mineração ou Aprendizado utilizados ou pela escassez dos dados disponíveis, os tornam computacionalmente caros ou imprecisos. Além disto, Sistemas de Recomendação em ambientes reais são dinâmicos, ou seja, os dados mudam com o passar do tempo seja com novas avaliações, novos usuários, novos itens ou mesmo atualizações de avaliações anteriores. A abordagem de Aprendizado Sem-Fim (SASF) visa um aprendizado autossupervisionado e autorreflexivo para, sobretudo, maximizar o aprendizado de um sistema com base em dados de fontes diversas, algoritmos que cooperem entre si para melhor modelar uma base de conhecimento e considerar a dinamicidade de problemas reais de aprendizado: Aprender amadurece com o tempo. Como já dito, sistemas de recomendação são dinâmicos e dependem de dados entre usuários e itens. Para minimizar esta dependência e prover resultados significativos e úteis aos usuários é apresentada neste trabalho uma abordagem de Sistema de Recomendação orientada pelos Princípios do Aprendizado Sem-Fim. Os resultados obtidos sugerem que é possível minimizar ou retardar a dependência de dados através de técnicas de acoplamento de classificadores e do controle do desvio de conceito. Com isto, é possível atuar com poucos dados de um sistema de recomendação que será dinâmico e receberá novas informações. Estas novas informações auxiliarão ainda mais no controle do desvio de conceito e na promoção de recomendações mais úteis. Por tudo isto, este trabalho apresenta como proposta o desenvolvimento de uma Abordagem para Sistemas de Recomendação baseada no Aprendizado Sem Fim, como forma de contribuir para o estado da arte em sistemas de recomendação e de implementar um sistema com resultados práticos através do Aprendizado sem Fim.