Relevance is an essential concept in information retrieval (IR) and relevance estimation is a fundamental IR task. It involves not only document relevance estimation, but also estimation of user's information need. Relevance-based language model aims to estimate a relevance model (i.e., a relevant query term distribution) from relevance feedback documents. The true relevance model should be generated from truly relevant documents. The ideal estimation of the true relevance model is expected to be not only effective in terms of mean retrieval performance (e.g., Mean Average Precision) over all the queries, but also stable in the sense that the performance is stable across different individual queries. In practice, however, in approximating/estimating the true relevance model, the improvement of retrieval effectiveness often sacrifices the retrieval stability, and vice versa. In this thesis, we propose to explore and analyze such effectiveness-stability tradeoff from a new perspective, i.e., the bias-variance tradeoff that is a fundamental theory in statistical estimation. We first formulate the bias, variance and the trade-off between them for retrieval performance as well as for query model estimation. We then analytically and empirically study a number of factors (e.g., query model complexity, query model combination, document weight smoothness and irrelevant documents removal) that can affect the bias and variance. Our study shows that the proposed bias-variance trade-off analysis can serve as an analytical framework for query model estimation. We then investigate in depth on two particular key factors: document weight smoothness and removal of irrelevant documents, in query model estimation, by proposing novel methods for document weight smoothing and irrelevance distribution separation, respectively. Systematic experimental evaluation on TREC collections shows that the proposed methods can improve both retrieval effectiveness and retrieval stability of query model estimation. In addition to the above main contributions, we also carry out initial exploration on two further directions: the formulation of bias-variance in personalization and looking at the query model estimation via a novel theoretical angle (i.e., Quantum theory) that has partially inspired our research.

Today’s increasing information supply raises the needfor more effective and automated information processing whereindividual information adaptation (personalization) is onepossible solution. Earlier computer systems for personalizationlacked the ability to easily define and measure theeffectiveness of personalization efforts. Numerous projectsfailed to live up to the their expectations, and the demand forevaluation increased. This thesis presents some underlying concepts and methodsfor implementing personalization in order to increase statedbusiness objectives. A personalization system was developedthat utilizes descriptions of information characteristics(metadata) to perform content based filtering in anon-intrusive way. Most of the described measurement methods forpersonalization in the literature are focused on improving theutility for the customer. The evaluation function of thepersonalization system described in this thesis takes thebusiness operator’s standpoint and pragmatically focuseson one or a few measurable business objectives. In order toverify operation of the personalization system, a functioncalled bifurcation was created. The bifurcation functiondivides the customers stochastically into two or morecontrolled groups with different personalizationconfigurations. Bygiving one of the controlled groups apersonalization configuration that deactivates thepersonalization, a reference group is created. The referencegroup is used to measure quantitatively objectives bycomparison with the groups with active personalization. Two different companies had their websites personalized andevaluated: one of Sweden’s largest recruitment servicesand the second largest Swedish daily newspaper. The purposewith the implementations was to define, measure, and increasethe business objectives. The results of the two case studiesshow that under propitious conditions, personalization can bemade to increase stated business objectives. Keywords:metadata, semantic web, personalization,information adaptation, one-to-one marketing, evaluation,optimization, personification, customization,individualization, internet, content filtering, automation. / <p>QCR 20161027</p>

This study explores the phenomenon of personalization within the political sphere. More specifically, it aims at exploring the strategical personalization of Swedish party leaders during the national election campaign of 2018. A further aim is to generate theoretical contributions to the area of personalization. As empirical material, the social media updates of three party leaders are analyzed under the period of 13 weeks before the election day, using a mixed methodological approach for an in-depth qualitative study. Regarding methodologies, grounded theory and coding is employed along with a rhetorical analysis that enable a systematic analysis of visual messages. The result indicates that the studied party leaders portrayed themselves as an integrated part of society, mediating a sense of presence and acknowledgement of the voters. In addition to this, the party leaders presented themselves as ordinary people and unified with the voters. The result also points to the practice of mediating a sense of social and cultural status, involving influencers in the communication and stepping into roles outside of the explicitly political arena. In terms of theoretical contributions, the generated theory contributes in the sense that it offers additional dimensions to the notion of "self-presentation", which is recognized as a common theme among prior theories. According to prior theories, individual politicians presents themselves as professional politicians and ordinary human beings. While the generated theory supports this perspective, it also highlights the aspect of intentionally integrating the voters into the self-presentations of individual politicians. Moreover, the generated theory emphasizes the perspective of social and cultural significance when regarding self-presentations of representatives.

La motivation principale de cette thèse est de proposer un système de recommandation personnalisé pour les plateformes d’informations. Pour cela, nous avons démontré que les opinions peuvent constituer un descripteur efficace pour améliorer la qualité de la recommandation. Au cours de cette thèse, nous avons abordé ce problème en proposant trois contributions principales. Tout d’abord, nous avons proposé un modèle de profil qui décrit avec précision les intérêts des utilisateurs ainsi que le contenu des articles de presse. Le modèle de profil proposé repose sur trois éléments : les entités nommées, les aspects et les sentiments. Nous avons testé notre modèle de profil sur les trois applications différentes que sont l’identification des orientations politiques des utilisateurs, la recommandation personnalisée des articles de presse et enfin la diversification de la liste des articles recommandés. Deuxièmement, nous avons proposé une approche de classement des opinions permettant de filtrer et sélectionner seulement les opinions pertinentes. Pour cela, nous avons utilisé une variation de la technique de PageRank pour définir le score de chaque opinion. Les résultats montrent que notre approche surpasse deux approches récemment proposées pour le classement des opinions. Troisièmement, nous avons étudié différentes façons d’enrichir le contenu des articles de presse par les opinions : par toutes les opinions, par seulement le topk des opinions, et enfin par un ensemble d’opinions diversifiées. Les résultats montrent que l’enrichissement des contenus des articles de presse / In this thesis, we have investigated how to exploit user-generated-content for personalized news recommendation purpose. The intuition behind this line of research is that the opinions provided by users, on news websites, represent a strong indicator about their profiles. We have addressed this problem by proposing three main contributions. Firstly, we have proposed a profile model that accurately describes both users’ interests and news article contents. The profile model was tested on three different applications ranging from identifying the political orientation of users to the context of news recommendation and the diversification of the list of recommended news articles. Results show that our profile model give much better results compared to state-of-the-art models. Secondly, we have investigated the problem of noise on opinions and how we can retrieve only relevant opinions in response to a given query.The proposed opinion ranking strategy is based on users’ debates features. We have used a variation of PageRank technique to define the score of each opinion. Results show that our approach outperforms two recent proposed opinions ranking strategies, particularly for controversial topics. Thirdly, we have investigated different ways of leveraging opinions on news article contents including all opinions, topk opinions based on opinion ranking strategy, and a set of diverse opinion. To extract a list of diverse opinions, we have employed a variation of an existing opinion diversification model. Results show that diverse opinions give the best performance over other leveraging strategies.

O trabalho Estação da Luz: desenho e realização de um painel para o saguão 2 de acesso às plataformas de trem mostra o percurso na criação do Painel Estação da Luz. Este painel, realizado pela autora deste trabalho, é composto por 14 módulos e mede 120 x 85 x 1 cm. Foi afixado nas paredes contíguas às escadas fixas do saguão 2, por intermédio das quais da área de integração entre as linhas da Companhia Paulista de Trens Metropolitanos (CPTM) e o Metrô paulistano se atinge a plataforma central da Estação da Luz. O objetivo do painel, além do estético, é travar uma relação imanente com a Estação da Luz, em cujo subsolo encontra-se instalado. Isso determinou a escolha de sua matéria-prima, ferro oxidado que remete ao material constituinte da estrutura de estação, ferro fundido , e de seu desenho. Este advém principalmente de sinais representativos da estação. Além de dialogar com a estação de trens, o painel interage com a arquitetura na qual está instalado, quanto ao partido adotado, a ambientação e os condicionantes locais. Assim, a análise desse objeto artístico insere-se no debate sobre as relações entre Arte e Arquitetura. A matéria-prima e a técnica para realização do painel também foram estudadas em função de questões estéticas. Neste sentido, foram feitos vários experimentos. / The work Estação da Luz: desenho e realização de um painel para o saguão 2 de acesso às plataformas de trem (Estação da Luz: design and realization of a panel for the Hall 2 accessing to railway platform) shows the trajectory of the Estação da Luz Panel creation by its author. The 120 x 85 x 1 cm panel is composed of 14 modules. It was installed on the walls at the sides of the Hall 2 fixed ladder that connects the Companhia Paulista de Trens Metropolitanos (CPTM) and São Paulo Metrô lines integration area and the central platform of Estação da Luz. The panel\'s goal, besides aesthetic, is to engage in an immanent relationship with Estação da Luz, beneath which it is installed. That determined the choice of oxidized iron as its raw material as a reference to the railway station material, cast iron , and of its design. The latter originated mainly from the station representative signs. Besides dialoguing with the railway station, the panel interacts with the architecture onto which it is installed, in terms of adopted proposal, ambience and local conditionings. So, the analysis of that artistic object inserts in the discussion about relationships between Art and Architecture. In the panel\'s realization, aesthetic issues also commanded the study of raw material and technique, and in this sense several essays were made.

O trabalho apresentado nesta tese trata do desenvolvimento de técnicas com suporte à ciência de contexto, baseadas nos padrões MPEG-4 e MPEG-7, para personalizar e adaptar conteúdo em TV Interativa. Um dos desafios dessa área é desenvolvimento de programas personalizados com rico conteúdo multimídia, com alta interatividade e que, além disso, sejam acessíveis a partir de uma variedade de dispositivos (fixos ou móveis), atendendo às expectativas de interação e de acesso dos usuários. Grande parte do problema está no fato de que os modos encontrados na literatura para representar, descrever e compor programas de TV Interativa não oferecem suporte a contexto, não permitem a separação entre descrições de programas e descrições de objetos e possuem baixa granulosidade de segmentação. Essas características dificultam e, em alguns casos, impedem o desenvolvimento de aplicações avançadas em TV Interativa. As técnicas desenvolvidas neste trabalho são baseadas em esquemas de descrição, compatíveis com o padrão MPEG-7, e na segmentação de programas em objetos MPEG-4. Os esquemas são utilizados para descrever a estrutura, a composição e a semântica de programas e de seus objetos componentes. Também foi definida e implantada uma infra-estrutura para produção, distribuição e consumo de programas. A utilização conjunta da infra-estrutura e das técnicas permite o desenvolvimento de aplicações avançadas em TV Interativa. Como um exemplo dessas aplicações, foi desenvolvido um serviço automático para personalizar e adaptar programas de TV Interativa, permitindo que um usuário possa acessar, sob demanda, programas especialmente produzidos para ele, contendo apenas assuntos de seu interesse e permitindo que o acesso possa ser realizado por dispositivos fixos ou móveis. / The work presented in this thesis developed techniques with context-awareness support, based on the MPEG-4 and MPEG-7 standards, in order to personalize and to adapt Interactive TV content. One of the challenges in this area is the development of personalized programs with rich multimedia content, high interactivity and accessibility through a variety of devices (mobile and non-mobile). Most part of the problem is that the approaches found in literature do not provide context support, do not allow separation between programs and objects descriptions and have low level of segmentation granularity. These features make difficult or impossible, in some cases, the development of Interactive TV applications. The techniques developed in this work are based on MPEG-7 compliant schemes and on programs segmentation into MPEG-4 objects. The schemes are used to describe structure, composition and semantics of programs and component objects. An infra-structure to creation, delivery and consumption of Interactive TV programs was also defined. The joint utilization of infra-structure and techniques allows for the development of Interactive TV advanced applications. As an example of these applications, this work developed an automatic Interactive TV personalization and adaptation service. This service allows a user to access, on-demand, a program specially designed to match his interests and allowing content access through devices with mobile and non-mobile features.

A extração de metadados semânticos de vídeos digitais para uso em serviços de personalização é importante, já que o conteúdo é adaptado segundo as preferências de cada usuário. Entretanto, apesar de serem encontradas várias propostas na literatura, as técnicas de indexação automática são capazes de gerar informações semânticas apenas quando o domínio do conteúdo é restrito. Alternativamente, existem técnicas para a criação manual dessas informações por profissionais, contudo, são dispendiosas e suscetíveis a erros. Uma possível solução seria explorar anotações colaborativas dos usuários, mas tal estratégia provoca a perda de individualidade dos dados, impedindo a extração de preferências do indivíduo a partir da interação. Este trabalho tem como objetivo propor uma arquitetura de personalização que permite a indexação multimídia de modo irrestrito e barato, utilizando anotações colaborativas, mas mantendo-se a individualidade dos dados para complementar o perfil de interesses do usuário com conceitos relevantes. A multimodalidade de metadados e de preferências também é explorada na presente tese, fornecendo maior robustez na extração dessas informações, e obtendo-se uma maior carga semântica que traz benefícios às aplicações. Como prova de conceito, este trabalho apresenta dois serviços de personalização que exploram a arquitetura proposta, avaliando os resultados por meio de comparações com abordagens previamente propostas na literatura / The extraction of semantic information from digital video is important to be used on personalization services because the content is adapted according to each users preferences. However, although it is possible to find several approaches in the literature, automatic indexing techniques are able to generate semantic metadata only when the contents domain is restricted. Alternatively, this information can be created manually by professionals, but this activity is time-consuming and error-prone. A possible solution would be to explore collaborative users annotations, but such approach has the disadvantage of lacking the individuality of annotations, hampering the extraction of users preferences from the interaction. This work has the objective of proposing a generic personalization architecture that allows multimedia indexing procedures to be accomplished in a cheap and unrestricted way. Such architecture uses collaborative annotations, but keeps the individuality of the data in order to augment the users profile with relevant concepts. The multimodality of metadata and users preferences is also explored in this work, which provides robustness during the extraction of semantic information, bringing benefits to applications. This work also presents two personalization services that explore the proposed architecture, along with evaluations that compare the obtained results with previously proposed approaches

Submitted by JOSIANE SANTOS DE OLIVEIRA (josianeso) on 2018-05-07T13:43:19Z No. of bitstreams: 1 Leonardo Dalmina_.pdf: 3273879 bytes, checksum: df157a5701b423e92352934d75e44473 (MD5) / Made available in DSpace on 2018-05-07T13:43:19Z (GMT). No. of bitstreams: 1 Leonardo Dalmina_.pdf: 3273879 bytes, checksum: df157a5701b423e92352934d75e44473 (MD5) Previous issue date: 2018-03-26 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O uso de elementos de design de jogos em contextos não relacionados a jogos, definido como gamificação, está sendo cada vez mais usado para aumentar a motivação e o engajamento dos usuários quando eles precisam executar uma tarefa em um ambiente não relacionado a jogo, como o local de trabalho, a escola ou uma aplicação de software. No entanto, quando a gamificação precisa ser implementada, um desafio enfrentado pelos desenvolvedores é identificar quais elementos do jogos engajarão efetivamente os usuários de um software com base em seus perfis de usuário e características motivacionais. Frequentemente, muitas pesquisas tendem a não incluir ou apenas apoiar os tipos de usuário e fatores motivacionais mais comuns. Em resposta a este desafio, esta dissertação propõe um modelo de gamificação genérico intitulado GamiProM que permite um desenvolvedor de software criar uma solução gamificada adaptativa para qualquer área fazendo uso de ontologias e regras, visando fornecer representação do conhecimento bem como adicionar um valor semântico à informação gerada pela gamificação e gerenciamento de perfil. O modelo é avaliado com um teste de correlação que identifica a existência de qualquer associação entre as necessidades psicológicas básicas dos usuários e suas motivações coletadas com a aplicação gamificada, desenvolvida para implementar o modelo proposto. Os resultados mostraram que as motivações coletadas dos perfis gamificados dos usuários têm uma correlação acima de 80% com as necessidades psicológicas básicas analisadas. / The use of game design elements in non-game contexts, defined as gamification, is being increasingly used to raise the motivation and engagement of users when they have to execute a task in a non-game environment, such as the workplace, the school or a software application. However, when gamification needs to be implemented, a challenge faced by developers is to identify what game elements will effectively engage the users of a software based on their user profiles and motivational characteristics. Often, many researches tend to not include or only support the most common user types and motivational factors. In response to this challenge, this thesis proposes a generic gamification model entitled GamiProM that allows a software developer to build an adaptive gamified solution for any area by making use of ontologies and rules, aiming to provide knowledge representation as well as add a semantic value to the information generated by gamification and profile management. The model is evaluated with a correlation test that identifies the existence of any association between the basic psychological needs of the users and their motivations collected with the gamified application, developed to implement the proposed model. The results showed that the motivations collected from the gamified profiles of the users have a correlation above 80% with the basic psychological needs analyzed.

La motivation principale de cette thèse est de proposer un système de recommandation personnalisé pour les plateformes d’informations. Pour cela, nous avons démontré que les opinions peuvent constituer un descripteur efficace pour améliorer la qualité de la recommandation. Au cours de cette thèse, nous avons abordé ce problème en proposant trois contributions principales. Tout d’abord, nous avons proposé un modèle de profil qui décrit avec précision les intérêts des utilisateurs ainsi que le contenu des articles de presse. Le modèle de profil proposé repose sur trois éléments : les entités nommées, les aspects et les sentiments. Nous avons testé notre modèle de profil sur les trois applications différentes que sont l’identification des orientations politiques des utilisateurs, la recommandation personnalisée des articles de presse et enfin la diversification de la liste des articles recommandés. Deuxièmement, nous avons proposé une approche de classement des opinions permettant de filtrer et sélectionner seulement les opinions pertinentes. Pour cela, nous avons utilisé une variation de la technique de PageRank pour définir le score de chaque opinion. Les résultats montrent que notre approche surpasse deux approches récemment proposées pour le classement des opinions. Troisièmement, nous avons étudié différentes façons d’enrichir le contenu des articles de presse par les opinions : par toutes les opinions, par seulement le topk des opinions, et enfin par un ensemble d’opinions diversifiées. Les résultats montrent que l’enrichissement des contenus des articles de presse / In this thesis, we have investigated how to exploit user-generated-content for personalized news recommendation purpose. The intuition behind this line of research is that the opinions provided by users, on news websites, represent a strong indicator about their profiles. We have addressed this problem by proposing three main contributions. Firstly, we have proposed a profile model that accurately describes both users’ interests and news article contents. The profile model was tested on three different applications ranging from identifying the political orientation of users to the context of news recommendation and the diversification of the list of recommended news articles. Results show that our profile model give much better results compared to state-of-the-art models. Secondly, we have investigated the problem of noise on opinions and how we can retrieve only relevant opinions in response to a given query.The proposed opinion ranking strategy is based on users’ debates features. We have used a variation of PageRank technique to define the score of each opinion. Results show that our approach outperforms two recent proposed opinions ranking strategies, particularly for controversial topics. Thirdly, we have investigated different ways of leveraging opinions on news article contents including all opinions, topk opinions based on opinion ranking strategy, and a set of diverse opinion. To extract a list of diverse opinions, we have employed a variation of an existing opinion diversification model. Results show that diverse opinions give the best performance over other leveraging strategies.

Denna studie handlar om individanpassning av informationsåtkomst och syftet med studien är att underlätta för utvecklare som ska ta fram nya, individanpassade lösningar för webben. Denna studie koncentreras till Skatteverkets webbplats. Studien försöker hitta svar på frågan om aktiv individanpassning gör att det blir lättare för användare att hitta vad de söker. En prototyp med en aktiv individanpassning har skapat och sedan jämförts med Skatteverkets webbplats. Detta har gjorts med hjälp av observationer och en tillämpning av tänka-högt-metoden. I resultatet av observationerna har det framkommit att även om respondenterna hittade vad de sökte på kortare tid när det använde Skatteverkets webbplats upplevdes prototypen som lättare att använda. Slutsatsen att en aktiv individanpassning gör det lättare för användaren att hitta rätt information har dragits. En förutsättning för att skapa en aktiv individanpassning är även att man tar hänsyn till användarens behov, kontextualisering, kategorisering och filtrering. / The subject of this essay is active personalization of information access. Its main goals are to make it easier for developers to create new, customized solutions for the web. The essay tries to find an answer to the question if personalization of information-access makes it easier for the user to find the information they need. The researchers developed a personalized-access prototype of the Swedish Tax Agency’s website and, through observations compared it with the original. The researchers applied the think-aloud-method in five observations documented as video recordings and observation schedules. In addition, a follow-up with a questionnaire was made after each observation in order to collect additional qualitative data. The observations showed that although the respondents found the information quickly at the Swedish Tax Agency’s website, they experienced the prototype as easier to use. The conclusion is that active personalization can make it easier for the users to find the right information, and that a prerequisite for creating an active personalization is that it takes better into account the user's needs by means of contextualization, categorization and filtering.