Online behavioural targeted advertising has become a leading method to increase theeffectiveness of advertisement online. The advertisement is based on information aboutusers’ internet activities collected by third party tracking services. However, there is a fineline between whether this should be considered a privacy leakage or an unharmful way toimprove the experience. This thesis presents a methodology for understanding and analyzing how significant personal information leakage to third parties is. To investigate theinformation leakage, a web crawler was implemented with the purpose to imitate fictiveusers browsing the web. The users’ activity online was based on the fictive user’s interestsand personal information. For every webpage visited, data such as text, URLs and screenshots were saved. The data were analyzed and the finding revealed that third parties generated targeted ads based on personal information regardless of the browser, user’s profile,and geographical location. However, we observed that targeted ads are a frequent practiceand noticed that categories valued more by advertisers are more intensely targeted.

abstract: In recent years, the rise in social media usage both vertically in terms of the number of users by platform and horizontally in terms of the number of platforms per user has led to data explosion. User-generated social media content provides an excellent opportunity to mine data of interest and to build resourceful applications. The rise in the number of healthcare-related social media platforms and the volume of healthcare knowledge available online in the last decade has resulted in increased social media usage for personal healthcare. In the United States, nearly ninety percent of adults, in the age group 50-75, have used social media to seek and share health information. Motivated by the growth of social media usage, this thesis focuses on healthcare-related applications, study various challenges posed by social media data, and address them through novel and effective machine learning algorithms. The major challenges for effectively and efficiently mining social media data to build functional applications include: (1) Data reliability and acceptance: most social media data (especially in the context of healthcare-related social media) is not regulated and little has been studied on the benefits of healthcare-specific social media; (2) Data heterogeneity: social media data is generated by users with both demographic and geographic diversity; (3) Model transparency and trustworthiness: most existing machine learning models for addressing heterogeneity are considered as black box models, not many providing explanations for why they do what they do to trust them. In response to these challenges, three main research directions have been investigated in this thesis: (1) Analyzing social media influence on healthcare: to study the real world impact of social media as a source to offer or seek support for patients with chronic health conditions; (2) Learning from task heterogeneity: to propose various models and algorithms that are adaptable to new social media platforms and robust to dynamic social media data, specifically on modeling user behaviors, identifying similar actors across platforms, and adapting black box models to a specific learning scenario; (3) Explaining heterogeneous models: to interpret predictive models in the presence of task heterogeneity. In this thesis, novel algorithms with theoretical analysis from various aspects (e.g., time complexity, convergence properties) have been proposed. The effectiveness and efficiency of the proposed algorithms is demonstrated by comparison with state-of-the-art methods and relevant case studies. / Dissertation/Thesis / Doctoral Dissertation Computer Science 2019

The focus of this dissertation is on creating Adaptive Intelligent User Interfaces to facilitate enhanced natural communication during the Human-Computer Interaction by recognizing users' affective states (i.e., emotions experienced by the users) and responding to those emotions by adapting to the current situation via an affective user model created for each user. Controlled experiments were designed and conducted in a laboratory environment and in a Virtual Reality environment to collect physiological data signals from participants experiencing specific emotions. Algorithms (k-Nearest Neighbor [KNN], Discriminant Function Analysis [DFA], Marquardt-Backpropagation [MBP], and Resilient Backpropagation [RBP]) were implemented to analyze the collected data signals and to find unique physiological patterns of emotions. Emotion Elicitation with Movie Clips Experiment was conducted to elicit Sadness, Anger, Surprise, Fear, Frustration, and Amusement from participants. Overall, the three algorithms: KNN, DFA, and MBP, could recognize emotions with 72.3%, 75.0%, and 84.1% accuracy, respectively. Driving Simulator experiment was conducted to elicit driving-related emotions and states (panic/fear, frustration/anger, and boredom/sleepiness). The KNN, MBP and RBP Algorithms were used to classify the physiological signals by corresponding emotions. Overall, KNN could classify these three emotions with 66.3%, MBP could classify them with 76.7% and RBP could classify them with 91.9% accuracy. Adaptation of the interface was designed to provide multi-modal feedback to the users about their current affective state and to respond to users' negative emotional states in order to decrease the possible negative impacts of those emotions. Bayesian Belief Networks formalization was employed to develop the User Model to enable the intelligent system to appropriately adapt to the current context and situation by considering user-dependent factors, such as: personality traits and preferences.

No description available.

Research on personalization, including recommender systems, focuses on applications such as in online shopping malls and simple information systems. These systems consider user profile and item information obtained from data explicitly entered by users. There it is possible to classify items involved and to personalize based on a direct mapping from user or user group to item or item group. However, in complex, dynamic, and professional information systems, such as digital libraries, additional capabilities are needed to achieve personalization to support their distinctive features: large numbers of digital objects, dynamic updates, sparse rating data, biased rating data on specific items, and challenges in getting explicit rating data from users. For this reason, more research on implicit rating data is recommended, because it is easy to obtain, suffers less from terminology issues, is more informative, and contains more user-centered information. In previous reports on my doctoral work, I discussed collecting, storing, processing, and utilizing implicit rating data of digital libraries for analysis and decision support. This dissertation presents a visualization tool, VUDM (Visual User-model Data Mining tool), utilizing implicit rating data, to demonstrate the effectiveness of implicit rating data in characterizing users, user communities, and usage trends of digital libraries. The results of user studies, performed both with typical end-users and with library experts, to test the usefulness of VUDM, support that implicit rating data is useful and can be utilized for digital library analysis software, so that both end users and experts can benefit. / Ph. D.

Author profiling is a field within Natural Language Processing (NLP) that is concerned with identifying various characteristics and demographic factors of authors, such as gender, age, location, native language, political orientation, and personality by analyzing the style and content of their writings. There is a growing interest in author profiling, with applications in marketing and advertising, opinion mining, personalization, recommendation systems, forensics, security, and defense. In this work, we build several classification models using NLP, Deep Learning, and classical Machine Learning techniques that can identify the gender and age of a Twitter user based on the textual contents of their correspondence (tweets) on the platform. Our SVM gender classifier utilizes a combination of word and character n-grams as features, dimensionality reduction using Latent Semantic Analysis (LSA), and a Support Vector Machine (SVM) classifier with linear kernel. At the PAN 2018 author profiling shared task, this model achieved the highest performance with 82.21%, 82.00%, and 80.90% accuracy on the English, Spanish, and Arabic datasets, respectively. Our age classifier was trained on a dataset of 11,160 Twitter users, using the same approach, though the age classification experiments are preliminary. Our Deep Learning gender classifiers are trained and tested on English datasets. Our feedforward neural network consisting of a word embedding layer, flattening, and two densely-connected layers achieves 79.57% accuracy, and our bidirectional Long Short-Term Memory (LSTM) neural network achieves 76.85% accuracy on the gender classification task.

A acomodação da diversidade de perfil de usuários no projeto de interface de sistemas é um problema freqüente nas atividades do projetista da interação homem-computador. Conseqüentemente, o usuário encontra barreiras ao utilizar interfaces que não foram produzidas para o seu perfil. Este trabalho apresenta uma solução, destinada aos projetistas de famílias de sistemas interativos, para a acomodação e expressão da diversidade por meio da criação e uso de padrões de interface em camadas de personas - as PICAPs. Neste conceito, os padrões de interface apóiam o projetista no direcionamento de soluções, pois representam um problema recorrente e uma solução abstrata para o problema, de tal modo que esta solução pode ser aplicada em várias instâncias diferentes do mesmo problema. As personas apóiam a caracterização dos perfis dos usuários que são foco do projeto de interface, possibilitando que o projetista aplique soluções de interface de acordo com o usuário. O conceito foi aplicado no contexto de governo eletrônico, pois tais sistemas devem ser usáveis por todos, em distinção de qualquer natureza, sendo assim um exemplo da necessidade de se considerar a diversidade. Uma pesquisa com 25 projetistas foi feita para avaliação da aplicabilidade do conceito. / Accommodating users\' profile diversity in systems interface projects is a frequent problem for the human computer interface designer. Therefore, his/her user is faced with barriers in the use of interfaces which were not designed for his/her profile. This work presents a solution for expressing and accommodating users\' diversity, which is useful for the HCI designer, especially for those who design families of products. PICAPS are interface design patterns with layers indexed by personas. The interface design patterns support the designer in employing proven solutions, for they represent a recurrent problem and its abstract solution in such a way that this solution can be applied to different instantiations of the same problem. PICAPs are structured in multiple layers to make possible the users\' diversity accommodation. The layers are indexed by personas as user\'s characterization resource. This concept has been applied to electronic government services, because such systems should be usable by any citizen and therefore are a good example of the user diversity problem. A field research with 25 designers has been performed to check the concept´s applicability.

L’avènement des Technologies de l’Information et de la Communication (TIC) représente une véritable opportunité dans la diffusion des ressources numériques et des connaissances pour tous. De nombreux résultats ont déjà été obtenus dans le cadre de l’étude de plates-formes informatiques numériques. Elles visent essentiellement à faciliter la mise à disposition des contenus numériques, accompagner l’utilisateur en lui offrant des modalités d’interaction avec le système et dans un contexte de formation, valider les connaissances acquises. Le déploiement large et massif des systèmes d’information de gestion de contenus numériques constitue une première génération d’outils de gestion. Dans ce cadre, des travaux de thèse ont été réalisés au sein du laboratoire L3I dans l’objectif de permettre à ces environnements de gérer efficacement l’interactivité avec leurs utilisateurs et d’adapter leurs exécutions en fonction des profils et du contexte de leurs utilisateurs. Certains résultats de ces travaux ont été mis en œuvre sur la plate-forme POLARIS du laboratoire. Ce contexte a créé un climat qui favorise la production massive et hétérogène de documents numériques dont la gestion (la conception, l’organisation, la sélection et l’usage) pose de nouveaux problèmes. L’objectif général de la thèse consiste à proposer des solutions permettant aux utilisateurs des applications interactives, de gérer leurs documents numériques afin d’éviter qu’ils ne soient submergés par l’immense quantité de documents qu’ils produisent et utilisent quotidiennement. En effet, pour mieux gérer ces documents, nous les encapsulons dans des structures complexes que nous appelons « ressources numériques ». Cela nous permet de mettre en œuvre nos mécanismes d’adaptation : pré-sélectionner les ressources numériques les plus adaptées aux activités de l’utilisateur et adapter les ressources numériques sélectionnées en fonction des caractéristiques de l’utilisateur et de son activité. Les contributions des travaux réalisés dans cette thèse se décrivent sur quatre niveaux. Premièrement, nous proposons un modèle de représentation des données qui caractérisent un utilisateur dans un système de gestion de ressources numériques. Deuxièmement, nous proposons des algorithmes et des mécanismes pour permettre à un utilisateur de sélectionner les ressources numériques pertinentes lors de la réalisation d’une activité. Les mécanismes de sélection proposés permettent une gestion consolidée des ressources numériques, parce qu’ils tiennent compte de la cohérence entre les ressources de l’activité afin de permettre une meilleure utilisation et maintenir une exécution cohérente de son activité. Troisièmement, nous proposons un modèle de ressources numériques qui s’adapte en fonction des caractéristiques de l’utilisateur et de l’activité. Quatrièmement, nous proposons un modèle d’architecture logicielle pour la conception d’une application interactive à exécution adaptative centrée sur la gestion des ressources numériques. Pour expérimenter et valider nos propositions, nous avons développé un prototype logiciel d’une plate-forme pour la gestion personnalisée et collaborative de ressources numériques dédié aux chercheurs appelé PRISE (PeRsonal Interactive Research Smart Environment). / The emergence of Information and Communication Technology (ICT) represents a real opportunity for spreading digital resources and knowledge for all. Many results have already been achieved through the study of digital computing environments. They primarily aim at facilitating the provision of digital content, supporting users by ensuring interaction means with the system, and validating the knowledge acquired. The broad and massive deployment of digital content management information systems has left its mark as first generation of digital document management tools. In this context, PhD thesis researches have been carried out in the laboratory L3i to enable these environments to manage interactivity with their users and adapt their performances to users’ profiles and usage contexts. Some results of these works have been implemented within the laboratory platform called POLARIS. This latter has created a climate that fosters the production of massive heterogeneous digital resources which their management (organization, selection, and use) raises new challenges. The main aim of the thesis is to propose solutions that enable users of interactive applications to manage their digital documents in such a way that they are not overwhelmed by the large amount of material they produce and use everyday. In other terms, to better manage these documents, we encapsulate them in complex structures named digital resources. That allows us to implement our adaptation mechanisms : to pre-select the most relevant digital resources to the user’s activities and adapt the selected digital resources to both user and activity characteristics.The contributions of our thesis can be described in four levels. First of all, we propose a data representation model that characterizes a user in a digital resources management system. Secondly, we propose algorithms and mechanisms to enable a user to select relevant digital resources when performing an activity. Selection mechanisms we proposed allow consolidated management of digital resources, since they take into account the consistency among the resources being used in a given activity to enable better use and maintain coherent execution of the activity. Thirdly, we propose a model for digital resources that adapts itself based on both user and activity characteristics. Finally, we propose a software architecture model for the design of an interactive application with adaptive execution centered on digital resource management. To experiment and validate our proposals, we developed a prototype of a platform for personalized and collaborative management of digital resources dedicated to researchers called PRISE (PeRsonal Interactive Research Smart Environment).

A acomodação da diversidade de perfil de usuários no projeto de interface de sistemas é um problema freqüente nas atividades do projetista da interação homem-computador. Conseqüentemente, o usuário encontra barreiras ao utilizar interfaces que não foram produzidas para o seu perfil. Este trabalho apresenta uma solução, destinada aos projetistas de famílias de sistemas interativos, para a acomodação e expressão da diversidade por meio da criação e uso de padrões de interface em camadas de personas - as PICAPs. Neste conceito, os padrões de interface apóiam o projetista no direcionamento de soluções, pois representam um problema recorrente e uma solução abstrata para o problema, de tal modo que esta solução pode ser aplicada em várias instâncias diferentes do mesmo problema. As personas apóiam a caracterização dos perfis dos usuários que são foco do projeto de interface, possibilitando que o projetista aplique soluções de interface de acordo com o usuário. O conceito foi aplicado no contexto de governo eletrônico, pois tais sistemas devem ser usáveis por todos, em distinção de qualquer natureza, sendo assim um exemplo da necessidade de se considerar a diversidade. Uma pesquisa com 25 projetistas foi feita para avaliação da aplicabilidade do conceito. / Accommodating users\' profile diversity in systems interface projects is a frequent problem for the human computer interface designer. Therefore, his/her user is faced with barriers in the use of interfaces which were not designed for his/her profile. This work presents a solution for expressing and accommodating users\' diversity, which is useful for the HCI designer, especially for those who design families of products. PICAPS are interface design patterns with layers indexed by personas. The interface design patterns support the designer in employing proven solutions, for they represent a recurrent problem and its abstract solution in such a way that this solution can be applied to different instantiations of the same problem. PICAPs are structured in multiple layers to make possible the users\' diversity accommodation. The layers are indexed by personas as user\'s characterization resource. This concept has been applied to electronic government services, because such systems should be usable by any citizen and therefore are a good example of the user diversity problem. A field research with 25 designers has been performed to check the concept´s applicability.

As the Internet became widespread all over the world, people started to share great amount of data on the web and almost every people joined different data networks in order to have a quick access to data shared among people and survive against the information overload on the web. Recommender systems are created to provide users more personalized information services and to make data available for people without an extra effort. Most of these systems aim to get or learn user preferences, explicitly or implicitly depending to the system, and guess &ldquo / preferable data&rdquo / that has not already been consumed by the user. Traditional approaches use user/item similarity or item content information to filter items for the active user / however most of the recent approaches also consider the trustworthiness of users. By using trustworthiness, only reliable users according to the target user opinion will be considered during information retrieval. Within this thesis work, a content boosted method of using trust data in recommender systems is proposed. It is aimed to be shown that people who trust the active user and the people, whom the active user trusts, also have correlated opinions with the active user. This results the fact that the rated items by these people can also be used while offering new items to users. For this research, www.epinions.com site is crawled, in order to access user trust relationships, product content information and review ratings which are ratings given by users to product reviews that are written by other users.