• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 85
  • 15
  • 8
  • 7
  • 6
  • 4
  • 4
  • 3
  • 3
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • Tagged with
  • 145
  • 145
  • 41
  • 34
  • 28
  • 24
  • 24
  • 23
  • 22
  • 22
  • 21
  • 20
  • 16
  • 15
  • 15
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
121

Personalized question-based cybersecurity recommendation systems

Moukala Both, Suzy Edith 08 1900 (has links)
En ces temps de pandémie Covid19, une énorme quantité de l’activité humaine est modifiée pour se faire à distance, notamment par des moyens électroniques. Cela rend plusieurs personnes et services vulnérables aux cyberattaques, d’où le besoin d’une éducation généralisée ou du moins accessible sur la cybersécurité. De nombreux efforts sont entrepris par les chercheurs, le gouvernement et les entreprises pour protéger et assurer la sécurité des individus contre les pirates et les cybercriminels. En raison du rôle important joué par les systèmes de recommandation dans la vie quotidienne de l'utilisateur, il est intéressant de voir comment nous pouvons combiner les systèmes de cybersécurité et de recommandation en tant que solutions alternatives pour aider les utilisateurs à comprendre les cyberattaques auxquelles ils peuvent être confrontés. Les systèmes de recommandation sont couramment utilisés par le commerce électronique, les réseaux sociaux et les plateformes de voyage, et ils sont basés sur des techniques de systèmes de recommandation traditionnels. Au vu des faits mentionnés ci-dessus, et le besoin de protéger les internautes, il devient important de fournir un système personnalisé, qui permet de partager les problèmes, d'interagir avec un système et de trouver des recommandations. Pour cela, ce travail propose « Cyberhelper », un système de recommandation de cybersécurité personnalisé basé sur des questions pour la sensibilisation à la cybersécurité. De plus, la plateforme proposée est équipée d'un algorithme hybride associé à trois différents algorithmes basés sur la connaissance, les utilisateurs et le contenu qui garantit une recommandation personnalisée optimale en fonction du modèle utilisateur et du contexte. Les résultats expérimentaux montrent que la précision obtenue en appliquant l'algorithme proposé est bien supérieure à la précision obtenue en utilisant d'autres mécanismes de système de recommandation traditionnels. Les résultats suggèrent également qu'en adoptant l'approche proposée, chaque utilisateur peut avoir une expérience utilisateur unique, ce qui peut l'aider à comprendre l'environnement de cybersécurité. / With the proliferation of the virtual universe and the multitude of services provided by the World Wide Web, a major concern arises: Security and privacy have never been more in jeopardy. Nowadays, with the Covid 19 pandemic, the world faces a new reality that pushed the majority of the workforce to telecommute. This thereby creates new vulnerabilities for cyber attackers to exploit. It’s important now more than ever, to educate and offer guidance towards good cybersecurity hygiene. In this context, a major effort has been dedicated by researchers, governments, and businesses alike to protect people online against hackers and cybercriminals. With a focus on strengthening the weakest link in the cybersecurity chain which is the human being, educational and awareness-raising tools have been put to use. However, most researchers focus on the “one size fits all” solutions which do not focus on the intricacies of individuals. This work aims to overcome that by contributing a personalized question-based recommender system. Named “Cyberhelper”, this work benefits from an existing mature body of research on recommender system algorithms along with recent research on non-user-specific question-based recommenders. The reported proof of concept holds potential for future work in adapting Cyberhelper as an everyday assistant for different types of users and different contexts.
122

[pt] VISUALIZANDO FATOS DE DADOS: UM ESTUDO COMPARATIVO DAS TÉCNICAS DE ANOTAÇÃO E SEU IMPACTO SOBRE AS PERCEPÇÕES DOS USUÁRIOS / [en] VISUALIZING DATA FACTS: A COMPARATIVE STUDY OF ANNOTATION TECHNIQUES AND THEIR IMPACT ON USERS PERCEPTIONS

DIEINISON JACK FREIRE BRAGA 03 July 2023 (has links)
[pt] Um número crescente de sistemas de visualização tem sido desenvolvido tanto comercialmente quanto na comunidade de pesquisa. Embora estas ferramentas possam ajudar na construção de gráficos, elas apresentam desafios para analistas não especialistas. Um desafio em particular é o de prover suporte para destacar visualmente fatos de dados em gráficos. O esforço empregado por analistas não especialistas ou designers (sem conhecimento de programação) para realizar anotações visuais pode ser complexo e demorado. Nesta pesquisa, investigamos representações visuais de fatos de dados para apoiar analistas não especialistas na exploração e comunicação de insights através dos dados. Para endereçar estes desafios, nós tornamos operacional um modelo conceitual que relaciona visualizações, fatos de dados e suas representações visuais. Implementamos o modelo em uma ferramenta de visualização chamada VisStoryMaker, que permite gerar gráficos anotados sem exigir conhecimento especializado. Para avaliar o seu valor percebido, conduzimos um estudo de métodos mistos com usuário comparando com o Tableau Public. No geral, a VisStoryMaker oferece uma abordagem fácil de usar para destacar visualmente fatos sobre dados, e o uso de anotações visuais de fatos sobre dados nas visualizações podem apoiar usuários não especialistas na exploração e comunicação por meio de dados. Entretanto, seu uso deve ser cuidadosamente considerado para evitar poluir visualmente os gráficos. / [en] A growing number of visualization systems have been developed both commercially and within the research community. While these tools can aid in building charts, they can also present challenges for non-expert analysts. One particular challenge is providing support to visually highlight data facts in graphs. The manual effort employed by non-expert analysts or designers (without programming skills) to create annotations can be complex and time-consuming. In this research, we investigate visual representations of data facts in supporting non-expert analysts to explore and communicate insights through data. To address these challenges, we developed a conceptual model relating visualizations, data facts, and their visual representations. We implemented it into a visualization tool named VisStoryMaker, which allows generating annotated charts without requiring specialized knowledge. To benchmark its perceived value, we conducted a mixed-methods user study comparing it to Tableau Public. Overall, VisStoryMaker provides an easy-to-use approach to highlight facts visually, and the use of visual annotations in data visualizations can support non-expert users in data exploration and communication. However, their use must be carefully considered and designed to avoid visually cluttering the charts.
123

Recommending in an Enterprise Social Media Stream without Explicit User Feedback

Lunze, Torsten, Katz, Philipp, Röhrborn, Dirk, Schill, Alexander January 2013 (has links)
Social Media Streams allow users to share user-generated content as well as aggregate different streams into one single stream. Additional Enterprise Social Media Streams organize the stream messages into projects with different usage patterns compared to public collaboration platforms such as Twitter. The aggregated stream helps the user to access the information in one single place but also leads to an information overload. Here, a recommendation engine can help to distinguish between relevant and irrelevant information for the users. In previous work we showed how features inferred from messages can predict relevant information and can be used to learn a user model. In this paper we show how this approach can be used in a productive enterprise social media stream application without using explicit user feedback. We develop a time binned evaluation measure which suits the scenario to steadily recommend messages of the stream. Finally, we evaluate our algorithm in different variations and show that it helps to identify relevant messages.
124

Developing a content and knowledge-based journal recommender system comparing distinct subject domains

Wijewickrema, Manjula 04 July 2019 (has links)
Die Aufgabe, ein passendes Journal zu finden, ist auf Grund von verschiedenen Einschränkungen nicht von Hand zu erledigen. Um also diese Problematik zu behandeln, entwickelt die aktuelle Untersuchung ein Journal-Empfehlungssystem, das – in einer Komponente – die inhaltlichen Ähnlichkeiten zwischen einem Manuskript und den existierenden Zeitschriftenartikeln in einem Korpus vergleicht. Das stellt die inhaltsbasierte Empfehlungskomponente des Systems dar. Zusätzlich beinhaltet das System eine wissensbasierte Empfehlungskomponente, um die Anforderungen des Autors bezüglich der Veröffentlichung auf Basis von 15 Journal-Auswahlkriterien zu berücksichtigen. Das neue System gibt Empfehlungen aus den im Directory of Open Access Journals indizierten Journals für zwei verschiedene Themengebiete: Medizin und Sozialwissenschaften. Die Ergebnisse zeigen, dass die Autoren aus den Themengebieten Medizin und Sozialwissenschaften mit den Empfehlungen des Systems zu 66,2% bzw. 58,8% einverstanden waren. Darüber hinaus wurde 35,5% der Autoren aus dem Bereich Medizin und 40,4% der Autoren aus den Sozialwissenschaften ein oder mehrere Journal(s) vorgeschlagen, das bzw. die für die Publikation besser geeignet war(en) als das Journal, in dem sie den Artikel veröffentlich hatten. Die durchschnittliche Leistung des Systems zeigte eine Abnahme von 15% in Medizin bzw. 18% in Sozialwissenschaften verglichen mit den gleichen Empfehlungen bei einer optimalen Sortierung. Leistungsverluste von 22,4% im Fach Medizin und 28,4% in den Sozialwissenschaften ergaben sich, wenn die durchschnittliche Leistung mit einem System verglichen wurde, das geeignete Empfehlungen für die 10 besten Resultate in der optimalen Reihenfolge sortiert abruft. Die vom Hybrid-Modell Empfehlungen zeigen zwar eine etwas bessere Leistung als die inhaltsbasierte Komponente, die Verbesserung war aber nicht statistisch signifikant. / The task of finding appropriate journals cannot be accomplished manually due to a number of limitations of the approach. Therefore, to address this issue, the current research develops a journal recommender system with two components: the first component compares the content similarities between a manuscript and the existing journal articles in a corpus. This represents the content-based recommender component of the system. In addition, the system includes a knowledge-based recommender component to consider authors’ publication requirements based on 15 journal selection factors. The new system makes recommendations from the open access journals indexed in the directory of open access journals for two distinct subject domains, namely medicine and social sciences. The results indicated that the authors from medicine and social sciences agree with the recommender’s suggestions by 66.2% and 58.8% respectively. Moreover, 35.5% of medicine and 40.4% of social sciences authors were suggested more appropriate journal(s) than the journal they already published in. Average performance of the system demonstrated 15% and 18% performance loss in medicine and social sciences respectively against the same suggestions after arranging according to the most appropriate order. Numbers were reported as 22.4% and 28.4% of loss in medicine and social sciences respectively when the average performance was compared with a system that retrieves appropriate suggestions for all 10 topmost results according to the most appropriate order. Although the hybrid recommender demonstrated a slight advancement of performance than the content-based component, the improvement was not statistically significant.
125

E-Fluence at the Point of Contact: Impact of Word-Of-Mouth and Personal Relevance of Services on Consumer Attitudes in Online Environments

Elias, Troy R.C. 03 September 2009 (has links)
No description available.
126

DRARS, a dynamic risk-aware recommender system / DRARS, un système de recommandation dynamique sensible au risque

Bouneffouf, Djallel 19 December 2013 (has links)
L’immense quantité d'information générée et gérée au quotidien par les systèmes d'information et leurs utilisateurs conduit inéluctablement à la problématique de surcharge d'information. Dans ce contexte, les systèmes de recommandation traditionnels fournissent des informations pertinentes aux utilisateurs. Néanmoins, avec la propagation récente des dispositifs mobiles (smartphones et tablettes), nous constatons une migration progressive des utilisateurs vers la manipulation d'environnements pervasifs. Le problème avec les approches de recommandation traditionnelles est qu'elles n'utilisent pas toute l'information disponible pour produire des recommandations. Davantage d’informations contextuelles pourraient être utilisées dans le processus de recommandation pour aboutir à des recommandations plus précises. Les systèmes de recommandation sensibles au contexte (CARS) combinent les caractéristiques des systèmes sensibles au contexte et des systèmes de recommandation afin de fournir des informations personnalisées aux utilisateurs dans des environnements ubiquitaires. Dans cette perspective où tout ce qui concerne l'utilisateur est dynamique, les contenus qu’il manipule et son environnement, deux questions principales doivent être adressées : i) Comment prendre en compte l'évolution des contenus de l’utilisateur? et ii) Comment éviter d’être intrusif, en particulier dans des situations critiques? En réponse à ces questions, nous avons développé un système de recommandation dynamique et sensible au risque appelé DRARS (Dynamic Risk-Aware Recommender System), qui modélise la recommandation sensible au contexte comme un problème de bandit. Ce système combine une technique de filtrage basée sur le contenu et un algorithme de bandit contextuel. Nous avons montré que DRARS améliore la stratégie de l'algorithme UCB (Upper Confidence Bound), le meilleur algorithme actuellement disponible, en calculant la valeur d'exploration la plus optimale pour maintenir un bon compromis entre exploration et exploitation basé sur le niveau de risque de la situation courante de l'utilisateur. Nous avons mené des expériences dans un contexte industriel avec des données réelles et des utilisateurs réels et nous avons montré que la prise en compte du niveau de risque de la situation de l'utilisateur augmentait significativement la performance du système de recommandation / The vast amount of information generated and maintained everyday by information systems and their users leads to the increasingly important concern of overload information. In this context, traditional recommender systems provide relevant information to the users. Nevertheless, with the recent dissemination of mobile devices (smartphones and tablets), there is a gradual user migration to the use of pervasive computing environments. The problem with the traditional recommendation approaches is that they do not utilize all available information for producing recommendations. More contextual parameters could be used in the recommendation process to result in more accurate recommendations. Context-Aware Recommender Systems (CARS) combine characteristics from context-aware systems and recommender systems in order to provide personalized recommendations to users in ubiquitous environments. In this perspective where everything about the user is dynamic, his/her content and his/her environment, two main issues have to be addressed: i) How to consider content evolution? and ii) How to avoid disturbing the user in risky situations?. In response to these problems, we have developed a dynamic risk sensitive recommendation system called DRARS (Dynamic Risk-Aware Recommender System), which model the context-aware recommendation as a bandit problem. This system combines a content-based technique and a contextual bandit algorithm. We have shown that DRARS improves the Upper Confidence Bound (UCB) policy, the currently available best algorithm, by calculating the most optimal exploration value to maintain a trade-off between exploration and exploitation based on the risk level of the current user's situation. We conducted experiments in an industrial context with real data and real users and we have shown that taking into account the risk level of users' situations significantly increases the performance of the recommender system
127

PROTECT_U: Un système communautaire pour la protection des usagers de Facebook

Gandouz, Ala Eddine 08 1900 (has links)
Article publié dans le journal « Journal of Information Security Research ». March 2012. / Chaque année, le nombre d’utilisateurs des réseaux sociaux augmente à une très grande vitesse. Des milliers de comptes usagés incluant des données privées sont créés quotidiennement. Un nombre incalculable de données privées et d'informations sensibles sont ainsi lues et partagées par les différents comptes. Ceci met en péril la vie privée et la sécurité de beaucoup d’utilisateurs de ces réseaux sociaux. Il est donc crucial de sensibiliser ces utilisateurs aux dangers potentiels qui les guettent. Nous présentons Protect_U (Hélou, Gandouz et al. 2012), un système de protection de la vie privée des utilisateurs de Facebook. Protect_U analyse le contenu des profils des utilisateurs et les classes selon quatre niveaux de risque : Low risk, medium risk, risky and critical. Il propose ensuite des recommandations personnalisées pour leur permettre de rendre leurs comptes plus sécuritaires. Pour ce faire, il fait appel à deux modèles de protection : local et communautaire. Le premier utilise les données personnelles de l’utilisateur afin de lui proposer des recommandations et le second recherche ses amis de confiance pour les inciter à participer à l’amélioration de la sécurité de son propre compte. / Social networking sites have experienced a steady and dramatic increase in the number of users over the past several years. Thousands of user accounts, each including a significant amount of private data, are created daily. As such, an almost countless amount of sensitive and private information is read and shared across the various accounts. This jeopardizes the privacy and safety of many social network users and mandates the need to increase the users’ awareness about the potential hazards they are exposed to on these sites. We introduce Protect_U (Hélou, Gandouz et al. 2012), a privacy protection system for Facebook users. Protect_U analyzes the content of user profiles and ranks them according to four risk levels: Low Risk, Medium Risk, Risky and Critical. The system then suggests personalized recommendations designed to allow users to increase the safety of their accounts. In order to achieve this, Protect_U draws upon both the local and community-based protection models. The first model uses a Facebook user’s personal data in order to suggest recommendations, and the second seeks out the user’s most trustworthy friends to encourage them to help improve the safety of his/her account.
128

Vers un meilleur accès aux informations pertinentes à l’aide du Web sémantique : application au domaine du e-tourisme / Towards a better access to relevant information with Semantic Web : application to the e-tourism domain

Lully, Vincent 17 December 2018 (has links)
Cette thèse part du constat qu’il y a une infobésité croissante sur le Web. Les deux types d’outils principaux, à savoir le système de recherche et celui de recommandation, qui sont conçus pour nous aider à explorer les données du Web, connaissent plusieurs problématiques dans : (1) l’assistance de la manifestation des besoins d’informations explicites, (2) la sélection des documents pertinents, et (3) la mise en valeur des documents sélectionnés. Nous proposons des approches mobilisant les technologies du Web sémantique afin de pallier à ces problématiques et d’améliorer l’accès aux informations pertinentes. Nous avons notamment proposé : (1) une approche sémantique d’auto-complétion qui aide les utilisateurs à formuler des requêtes de recherche plus longues et plus riches, (2) des approches de recommandation utilisant des liens hiérarchiques et transversaux des graphes de connaissances pour améliorer la pertinence, (3) un framework d’affinité sémantique pour intégrer des données sémantiques et sociales pour parvenir à des recommandations qualitativement équilibrées en termes de pertinence, diversité et nouveauté, (4) des approches sémantiques visant à améliorer la pertinence, l’intelligibilité et la convivialité des explications des recommandations, (5) deux approches de profilage sémantique utilisateur à partir des images, et (6) une approche de sélection des meilleures images pour accompagner les documents recommandés dans les bannières de recommandation. Nous avons implémenté et appliqué nos approches dans le domaine du e-tourisme. Elles ont été dûment évaluées quantitativement avec des jeux de données vérité terrain et qualitativement à travers des études utilisateurs. / This thesis starts with the observation that there is an increasing infobesity on the Web. The two main types of tools, namely the search engine and the recommender system, which are designed to help us explore the Web data, have several problems: (1) in helping users express their explicit information needs, (2) in selecting relevant documents, and (3) in valuing the selected documents. We propose several approaches using Semantic Web technologies to remedy these problems and to improve the access to relevant information. We propose particularly: (1) a semantic auto-completion approach which helps users formulate longer and richer search queries, (2) several recommendation approaches using the hierarchical and transversal links in knowledge graphs to improve the relevance of the recommendations, (3) a semantic affinity framework to integrate semantic and social data to yield qualitatively balanced recommendations in terms of relevance, diversity and novelty, (4) several recommendation explanation approaches aiming at improving the relevance, the intelligibility and the user-friendliness, (5) two image user profiling approaches and (6) an approach which selects the best images to accompany the recommended documents in recommendation banners. We implemented and applied our approaches in the e-tourism domain. They have been properly evaluated quantitatively with ground-truth datasets and qualitatively through user studies.
129

網路評比資料之統計分析 / Statistical analysis of online rating data

張孫浩 Unknown Date (has links)
隨著網路的發達,各式各樣的資訊和商品也在網路上充斥著,使用者尋找資訊或是上網購物時,有的網站有推薦系統(recommender system)能提供使用者相關資訊或商品。若推薦系統能夠讓消費者所搜尋的相關資訊或商品能夠符合他們的習性時,便能讓消費者增加對系統的信賴程度,因此系統是否能準確預測出使用者的偏好就成為一個重要的課題。本研究使用兩筆資料,並以相關研究的三篇文獻進行分析和比較。這三篇文獻分別為IRT模型法(IRT model-based method)、相關係數法(correlation-coefficient method)、以及矩陣分解法(matrix factorization)。 在經過一連串的實證分析後,歸納出以下結論: 1. 模型法在預測方面雖然精確度不如其他兩種方法來的好,但是模型有解釋變數之間的關係以及預測機率的圖表展示,因此這個方法仍有存在的價值。 2. 相關係數法容易因為評分稀疏性的問題而無法預測,建議可以搭配內容式推薦系統的運作方式協助推薦。 3. 矩陣分解法在預測上雖然比IRT模型法還好,但分量的數字只是一個最佳化的結果,實際上無法解釋這些分量和數字的意義。 / With the growth of the internet, websites are full of a variety of information and products. When users find the information or surf the internet to shopping, some websites provide users recommender system to find with which related. Hence, whether the recommender system can predict the users' preference is an important topic. This study used two data,which are "Mondo" and "MovieLens", and we used three related references to analyze and compare them. The three references are following: IRT model-based method, Correlation-coefficient method, and Matrix factorization. After the data analysis, we get the following conclusions: 1. IRT model-based method is worse then other methods in predicting, but it can explain the relationship of variables and display the graph of predicting probabilities. Hence this method still has it's value. 2. Correlation-coefficient method is hard to predict because of sparsity. We can connect it with content filtering approach. 3. Although matrix factorization is better then IRT model-based method in predicting, the vectors is a result of optimization. It may be hard to explain the meaning of the vectors.
130

Design of a Recommender System for Participatory Media Built on a Tetherless Communication Infrastructure

Seth, Aaditeshwar January 2008 (has links)
We address the challenge of providing low-cost, universal access of useful information to people in different parts of the globe. We achieve this by following two strategies. First, we focus on the delivery of information through computerized devices and prototype new methods for making that delivery possible in a secure, low-cost, and universal manner. Second, we focus on the use of participatory media, such as blogs, in the context of news related content, and develop methods to recommend useful information that will be of interest to users. To achieve the first goal, we have designed a low-cost wireless system for Internet access in rural areas, and a smartphone-based system for the opportunistic use of WiFi connectivity to reduce the cost of data transfer on multi-NIC mobile devices. Included is a methodology for secure communication using identity based cryptography. For the second goal of identifying useful information, we make use of sociological theories regarding social networks in mass-media to develop a model of how participatory media can offer users effective news-related information. We then use this model to design a recommender system for participatory media content that pushes useful information to people in a personalized fashion. Our algorithms provide an order of magnitude better performance in terms of recommendation accuracy than other state-of-the-art recommender systems. Our work provides some fundamental insights into the design of low-cost communication systems and the provision of useful messages to users in participatory media through a multi-disciplinary approach. The result is a framework that efficiently and effectively delivers information to people in remote corners of the world.

Page generated in 0.0913 seconds