A North American man has approximately a one in six chance of being diagnosed with prostate cancer in his lifetime. In most cases, there is no clearly optimal treatment, so he may be invited to participate in a treatment decision between several medically reasonable options, each with potential short- and long-term side effects. Information needs are high at diagnosis and can continue to be elevated for years or decades. Many men and their families seek information online, where, due partly to the array of websites available and high variation in information preferences, it can be difficult to find personally relevant and useful websites. This research sought to address this issue by developing methods to categorize prostate cancer websites and exploring quantitative and qualitative relationships between websites, information-seekers, and individuals’ assessments of websites. The research involved a series of three studies. In the first study, 29 men with prostate cancer participated in a needs assessment involving questionnaires, an interview, and interaction with a prototype website. In the second study, a detailed classification system was developed and applied to a set of forty websites selected to be representative of the variety of prostate cancer websites available. The third (online) study collected clinical, cognitive, and psychosocial details from 65 participants along with their ratings of websites from study two. A number of hypotheses were tested. One finding was that, compared to men with greater trust, men with lower trust in their physician tended to judge commercial websites as less relevant and useful, and found websites with descriptions of personal experiences more relevant and useful. Analyses also addressed a number of exploratory questions, including whether website and individual attributes might predict preferences for websites. Using discriminant analysis on 80% of the data, two functions were identified that predicted ratings significantly better than chance. These relationships were then validated with 20% of the data held back for testing. The results are discussed in terms of their implications for information tailoring and recommender systems for prostate cancer patients searching for information online. Limitations of the current research and recommendations for future research are also presented.

A North American man has approximately a one in six chance of being diagnosed with prostate cancer in his lifetime. In most cases, there is no clearly optimal treatment, so he may be invited to participate in a treatment decision between several medically reasonable options, each with potential short- and long-term side effects. Information needs are high at diagnosis and can continue to be elevated for years or decades. Many men and their families seek information online, where, due partly to the array of websites available and high variation in information preferences, it can be difficult to find personally relevant and useful websites. This research sought to address this issue by developing methods to categorize prostate cancer websites and exploring quantitative and qualitative relationships between websites, information-seekers, and individuals’ assessments of websites. The research involved a series of three studies. In the first study, 29 men with prostate cancer participated in a needs assessment involving questionnaires, an interview, and interaction with a prototype website. In the second study, a detailed classification system was developed and applied to a set of forty websites selected to be representative of the variety of prostate cancer websites available. The third (online) study collected clinical, cognitive, and psychosocial details from 65 participants along with their ratings of websites from study two. A number of hypotheses were tested. One finding was that, compared to men with greater trust, men with lower trust in their physician tended to judge commercial websites as less relevant and useful, and found websites with descriptions of personal experiences more relevant and useful. Analyses also addressed a number of exploratory questions, including whether website and individual attributes might predict preferences for websites. Using discriminant analysis on 80% of the data, two functions were identified that predicted ratings significantly better than chance. These relationships were then validated with 20% of the data held back for testing. The results are discussed in terms of their implications for information tailoring and recommender systems for prostate cancer patients searching for information online. Limitations of the current research and recommendations for future research are also presented.

With the increasing amount of data on web, people start to need tools which will help them to deal with the most significant ones among the thousands. The idea of a system which recommends items to its users emerged to fulfill this inevitable need. But most of the recommender systems make recommendations for individuals. On the other hand, some people need recommendation for items which they will use or for activities which they will attend together. Group recommenders serve for these purposes. Group recommenders diverge from individual recommenders such that they need to aggregate members of the group in a joint model, and in order to do so, they need a user satisfaction function. There are two different aggregation methods and a few different satisfaction functions for group recommendation process. Reading groups domain is a new domain for group recommenders. In this thesis we propose a web based group recommender system which is called BoRGo: Book Recommender for Reading Groups , for reading groups domain. BoRGo uses a new information filtering technique and present a media for post recommendation processes. We present comparative evaluation results of this new technique in this thesis.

Thesis (Ph.D.) -- Mississippi State University. Department of Management and Information Systems. / Title from title screen. Includes bibliographical references.

Thesis (MA)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: We are living in an increasingly global and connected society with information creation increasing at exponential rates. The research sets out to help solve the problem of mitigating the effects of information overload in order to increase the novelty of our interactions in the digital age. Online social-networks and microblogging services allow people across the world to take part in a public conversation. These tools have inherent constraints on how much communication can feasibly occur. Become too connected and a user will receive too much information to reasonably process. On Twitter (a microblogging service), lists are a tool for users to create separate feeds. The research determines whether lists are an effective tool for coping with information overload (abundance of updates). Using models of sustainable online discourse and information overload on computer-mediated communication tools, the research found that lists are an effective tool to cope with information overload on microblogging services. Quantitatively, individuals who make use of lists follow more users and when they start using lists they increase the amount of information resources (following other users) at a greater rate than those who do not use lists. Qualitatively, the research also provides insight into the reasons why people use lists. The research adds new academic relevance to ‘information overload’ and ‘online sustainability’ models previously not used in the context of feed-based online CMC tools, and deepens the understanding and importance of usercurated filtering as a way to reap the benefits from the increasing abundance of information in the digital age. / AFRIKAANSE OPSOMMING: Ons leef in ’n toenemend globale en gekonnekteerde samelewing waarin inligtingskepping toeneem teen ’n eksponensiële koers. Hierdie navorsing het ten doel om die newe-effekte van die oorvloed van inligting te verlig sodat daar meer waarde uit ons interaksies in die digitale era kan geput kan word. Aanlyn sosiale-netwerke en mikroblog-dienste laat mense wêreldwyd toe om deel te neem in ’n openbare gesprek. Hierdie aanlyn gereedskap het egter inherente beperkinge op hoeveel kommunikasie prakties moontlik is. Wanneer gebruikers té gekonnekteer raak, word daar te veel ingligting ontvang om redelikerwys verwerk te kan word. Op Twitter (’n mikroblog-diens) is lyste ’n hulpmiddel waarmee gebruikers afsonderlike strome van inligting kan skep. Deur die gebruik van modelle van ‘volhoubare aanlyn diskoers’ en ‘inligtingoorlading’, bewys hierdie navorsing dat lyste ’n doeltreffende hulpmiddel is om die oorvloed van inligting te verlig op mikroblog-dienste. Kwantitatief volg gebruikers wat lyste gebruik meer gebruikers vergeleke met die wat nie lyste gebruik nie. Wanner hul lyste begin gebruik, volg hulle gebruikers teen ’n hoër koers as dié wat nie lyste gebruik nie. Kwalitatief bied die navorsing ook insig oor die redes vir die gebruik van lyste. Die navorsing onderstreep die akademiese relevansie van ‘inligtingoorlading’ en ‘aanlyn volhoubaarheid’ modelle wat nie voorheen gebruik is in die konteks van stroom-gebaseerde aanlyn gereedskap nie, en verdiep die begrip en belangrikheid van gebruiker-saamgestelde filtrering as ’n manier om die voordele te trek uit die toenemende oorvloed van inligting in die digitale era.

Esta dissertação apresenta novos resultados para a solução de problemas de implementação computacional na estimativa de sistemas singulares e sistemas Markovianos. São apresentados algoritmos alternativos para problemas de filtragem de maneira a minimizar problemas causados principalmente por erros de arredondamento e mal condicionamento de matrizes. O trabalho envolve basicamente algoritmos array e filtragem de informação para a estimativa de sistemas singulares nominais e robustos. Também é deduzido um algoritmo array para a filtragem de sistemas lineares sujeitos a saltos Markovianos. / This dissertation presents new results to solve computational implementation problems to estimate singular and Markovian systems. Alternative algorithms to handle computational filtering errors due rounding errors and ill-conditioned matrices are developed. This dissertation comprehends basically array algorithms and information filters for the estimate of nominal and robust singular systems. Also, it is developed an array algorithm for Markovian jump linear systems filtering.

Orientador: Ricardo da Silva Torres / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-11T00:55:04Z (GMT). No. of bitstreams: 1 Pedronette_DanielCarlosGuimaraes_M.pdf: 7416929 bytes, checksum: be0fe860604d3545272fbe4bd4aaa8df (MD5) Previous issue date: 2008 / Resumo: Em virtude do crescimento acelerado de conteúdo nas mais diversas aplicações de bibliotecas digitais, a tarefa de localizar objetos digitais de interesse é cada vez mais desafiadora. Sob essa perspectiva, técnicas de recomendação procuram prover, de acordo com as preferências do usuário final, alternativas de escolha de objetos mantidos em uma biblioteca digital. Essa dissertação concentra-se em aspectos relacionados às técnicas de recomendação e suas interações com aplicações de bibliotecas digitais. Uma plataforma de serviços de recomendação, chamada RecS-DL, é proposta, visando ampliar as possibilidades de utilização das ferramentas de recomendação. A Plataforma RecS-DL apresentada é independente de domínio de aplicação, de tecnologias e técnicas de recomendação. O serviço de recomendação oferecido pode ser facilmente agregado a bibliotecas digitais clientes, assim como novos mecanismos de recomendação podem ser acoplados à plataforma de maneira dinâmica. Este trabalho também apresenta uma especificação formal da plataforma de serviços de recomendação proposta a partir do Arcabouço 5S. Para isso foram propostas novas definições e extensões de conceitos deste arcabouço. Por fim, são apresentados os resultados obtidos a partir de testes realizados com a plataforma. Experimentos foram conduzidos considerando bibliotecas digitais reais e avaliações por potenciais usuários. Resultados experimentais ratificam a hipótese de que a plataforma facilita a interoperabilidade de ferramentas de recomendação em bibliotecas digitais / Abstract: The increasing amount of data in the most diverse digital libraries applications makes the process of finding relevant digital objects a challenging task. From this perspective, recommendation techniques can provide, according to user preferences, relevant digital objects stored in a digital library.This dissertation focuses on recommendation techniques and their interactions with digital libraries applications. A platform for recommendation services, called RecS-DL, has been proposed to support the use of recommendation tools. The proposed RecS-DL Platform is independent of application domain, technology, and recommendation techniques. The recommendation services offered by the platform can be easily incorporated into digital libraries systems. Furthermore, new recommendation engines can also be plugged into the platform in a dynamic way. This work also presents a formal specification of the proposed platform, using the 5S Framework. To do this, new definitions and extensions of this framework are proposed. Finally, we present the results obtained from tests performed with the platform. Experiments were conducted considering real digital libraries and evaluations made by potential users. Experimental results confirm that the platform facilitates the interoperability of recommendation tools in digital libraries systems / Mestrado / Sistemas de Informação / Mestre em Ciência da Computação

No description available.

Submitted by Maria Aparecida (cidazen@gmail.com) on 2017-08-29T14:17:25Z No. of bitstreams: 1 Fabio Augusto.pdf: 2319314 bytes, checksum: 7dc99465ac724efe228c61bb9dfafa80 (MD5) / Made available in DSpace on 2017-08-29T14:17:25Z (GMT). No. of bitstreams: 1 Fabio Augusto.pdf: 2319314 bytes, checksum: 7dc99465ac724efe228c61bb9dfafa80 (MD5) Previous issue date: 2009-05-18 / Several techniques for extracting meaning from text in order to construct more accurate internal representations of both queries and information items in retrieval systems have been already proposed. However, there is a lack of semantic retrieval models to provide appropriate abstractions of these techniques. This work proposes a knowledge--based information retrieval model that explores the semantic content of information items . The internal representation of information items is based on user interest groups, called “semantic cases”. The model also defines a criteria for retrieve information items and a function for ordering the results that uses similarity measures based on semantic distance between semantic cases items. The model was instantiated by a sample system built upon the tributary legal domain using the specialization of the ONTOJURIS, a generic legal ontology, called ONTOTRIB. Legal normative instruments can be instantiated in a knowledge base by ONTOTRIB classes. The results obtained for this specific domain showed an improvement in the precision rates compared to a keyword-based system. / Várias técnicas para extrair significado de textos com o objetivo de construir representações internas mais precisas, tanto para itens de informação quanto para consultas em sistemas de recuperação já foram propostas. Contudo, faltam modelos de recuperação baseados em semântica que especifiquem abstrações apropriadas para essas técnicas. Este trabalho apresenta um modelo de recuperação baseado no conhecimento que explora o conteúdo semântico dos itens de informação. A representação interna dos itens de informação é baseada em grupos de interesse do usuário chamados de “casos semânticos”. O modelo também define um critério para a recuperação dos itens de informação e uma função para ordenar os resultados obtidos que utiliza medidas de similaridade baseadas na distância semântica entre os elementos das representações internas. O modelo foi instanciado em um sistema construído para o domínio jurídico tributário usando a ontologia ONTOTRIB, uma extensão da ontologia genérica ONTOJURIS, que permite a instanciação de instrumentos jurídico-tributários. Os resultados obtidos nos testes realizados neste domínio específico apontaram uma melhoria da precisão em relação a um sistema baseado em palavras-chave.

Over the last decade, excess amount of information is being provided on the web and information filtering systems such as recommender systems have become one of the most important technologies to overcome the &bdquo / Information Overload&amp / #8223 / problem by providing personalized services to users. Several researches have been made to improve quality of recommendations and provide maximum user satisfaction within a single domain based on the domain specific knowledge. However, the current infrastructures of the recommender systems cannot provide the complete mechanisms to meet user needs in several domains and recommender systems show poor performance in cross-domain item recommendations. Within this thesis work, a dynamic framework is proposed which differs from the previous works as it focuses on the easy development of knowledge-based recommenders and it proposes an intensive cross domain capability with the help of domain knowledge. The framework has a generic and flexible structure that data models and user interfaces are generated based on ontologies. New recommendation domains can be integrated to the framework easily in order to improve recommendation diversity. The cross-domain recommendation is accomplished via an abstraction in domain features if the direct matching of the domain features is not possible when the domains are not very close to each other.