Ces dernières années, le contenu disponible sur le Web a augmenté de manière considérable dans ce qu’on appelle communément le Web social. Pour l’utilisateur moyen, il devient de plus en plus difficile de recevoir du contenu de qualité sans se voir rapidement submergé par le flot incessant de publications. Pour les fournisseurs de service, le passage à l’échelle reste problématique. L’objectif de cette thèse est d’aboutir à une meilleure expérience utilisateur à travers la mise en place de systèmes de filtrage et de recommandation. Le filtrage consiste à offrir la possibilité à un utilisateur de ne recevoir qu’un sous ensemble des publications des comptes auxquels il est abonné. Tandis que la recommandation permet la découverte d’information à travers la suggestion de comptes à suivre sur des sujets donnés. Nous avons élaboré MicroFilter un système de filtrage passant à l’échelle capable de gérer des flux issus du Web ainsi que RecLand, un système de recommandation qui tire parti de la topologie du réseau ainsi que du contenu afin de générer des recommandations pertinentes. / In the last years, the amount of available data on the social Web has exploded. For the average user, it became hard to find quality content without being overwhelmed with publications. For service providers, the scalability of such services became a challenging task. The aim of this thesis is to achieve a better user experience by offering the filtering and recommendation features. Filtering consists to provide for a given user, the ability of receiving only a subset of the publications from the direct network. Where recommendation allows content discovery by suggesting relevant content producers on given topics. We developed MicroFilter, a scalable filtering system able to handle Web-like data flows and RecLand, a recommender system that takes advantage of the network topology as well as the content in order to provide relevant recommendations.

Ces dernières années, le contenu disponible sur le Web a augmenté de manière considérable dans ce qu’on appelle communément le Web social. Pour l’utilisateur moyen, il devient de plus en plus difficile de recevoir du contenu de qualité sans se voir rapidement submergé par le flot incessant de publications. Pour les fournisseurs de service, le passage à l’échelle reste problématique. L’objectif de cette thèse est d’aboutir à une meilleure expérience utilisateur à travers la mise en place de systèmes de filtrage et de recommandation. Le filtrage consiste à offrir la possibilité à un utilisateur de ne recevoir qu’un sous ensemble des publications des comptes auxquels il est abonné. Tandis que la recommandation permet la découverte d’information à travers la suggestion de comptes à suivre sur des sujets donnés. Nous avons élaboré MicroFilter un système de filtrage passant à l’échelle capable de gérer des flux issus du Web ainsi que RecLand, un système de recommandation qui tire parti de la topologie du réseau ainsi que du contenu afin de générer des recommandations pertinentes. / In the last years, the amount of available data on the social Web has exploded. For the average user, it became hard to find quality content without being overwhelmed with publications. For service providers, the scalability of such services became a challenging task. The aim of this thesis is to achieve a better user experience by offering the filtering and recommendation features. Filtering consists to provide for a given user, the ability of receiving only a subset of the publications from the direct network. Where recommendation allows content discovery by suggesting relevant content producers on given topics. We developed MicroFilter, a scalable filtering system able to handle Web-like data flows and RecLand, a recommender system that takes advantage of the network topology as well as the content in order to provide relevant recommendations.

Communication methods in the corporate environment must change. Organisations can no longer expect effective communication when using intranet sites or sending employees countless emails. This type of communication does not create the learning environment and most employees either do not read the intranet sites, or there is just too much email which could be seen as spam by the employee. These types of technologies also create a culture where organisations are lead from the top and there is not a culture or platform to create feedback loops. Most large corporate organisations have a tendency where many silos are created and cliques are formed, which is not in line with the culture of a learning organisation. Although there are a number of studies which look at how Internet based micro-blogging affect social connectedness, there is however limited information as to the effect that micro-blogging, if used internal to the organisation, would have on employee engagement, or how it can affect the competitive nature to the organisation. The research is exploratory in nature and set out to review what impact internal micro-blogging has on the organisation. The research uses Vodacom, one of the leading Information and communications technology (ICT) and telecommunications companies in South Africa, as the case study, as micro-blogging was recently introduced into this organisation.

Communication methods in the corporate environment must change. Organisations can no longer expect effective communication when using intranet sites or sending employees countless emails. This type of communication does not create the learning environment and most employees either do not read the intranet sites, or there is just too much email which could be seen as spam by the employee. These types of technologies also create a culture where organisations are lead from the top and there is not a culture or platform to create feedback loops. Most large corporate organisations have a tendency where many silos are created and cliques are formed, which is not in line with the culture of a learning organisation. Although there are a number of studies which look at how Internet based micro-blogging affect social connectedness, there is however limited information as to the effect that micro-blogging, if used internal to the organisation, would have on employee engagement, or how it can affect the competitive nature to the organisation. The research is exploratory in nature and set out to review what impact internal micro-blogging has on the organisation. The research uses Vodacom, one of the leading Information and communications technology (ICT) and telecommunications companies in South Africa, as the case study, as micro-blogging was recently introduced into this organisation.

With the rise of Web 2.0 concept, the development of virtual communities focuses more on the factors of relationship and interaction. Micro-blogging, the kind of community platforms that limit the length of a post and emphasize real-time and interactive features, has become popular these years. Seeing the potential of micro-blogging marketing, many businesses start to use micro-blogging as a promotional channel. With the different characteristics and multiple features of micro-blogging services, however, most businesses are still exploring how to make relationship marketing successful. This study uses commitment-trust theory by Morgan & Hunt (1994) to investigate the building of trust and commitment between users and communities and their influence on community loyalty from the perspective of relationship commitment. An online-survey was conducted among Plurk users, and 300 valid respondents were used for data analysis by structural equation modeling (SEM). The results indicate that commitment and trust have a significantly positive effect on loyalty. Trust is significantly influenced by shared values, communication and opportunistic behavior, and commitment is significantly influenced by relation-ship benefits and trust. Information quality and shared values, however, do not have significant impact on commitment. Furthermore, user characteristics and usage frequency have significant moderating effect on the level of causal influence. The results show the importance of trust and commitment building in micro-blogging and provide useful suggestions for micro-blogging marketing.

Submitted by Dilene Paulo (dilene.fatima@ufcg.edu.br) on 2017-12-01T15:44:03Z No. of bitstreams: 1 ANDRÉ LUIZ FIRMINO ALVES - DISSERTAÇÃO PPGCC 2014.pdf: 8491551 bytes, checksum: 9c049bdf4b551742efcd94bdc4c8c020 (MD5) / Made available in DSpace on 2017-12-01T15:44:03Z (GMT). No. of bitstreams: 1 ANDRÉ LUIZ FIRMINO ALVES - DISSERTAÇÃO PPGCC 2014.pdf: 8491551 bytes, checksum: 9c049bdf4b551742efcd94bdc4c8c020 (MD5) Previous issue date: 2014 / Capes / A proliferação dos meios de comunicação social na Web, tais como blogs, fóruns de discussões, sites de avaliação de produtos, microblogs e redes sociais, proporcionou um volume de dados opinativos armazenados em formato digital nunca visto na história da humanidade. Esta quantidade de dados, em sua grande maioria não estruturados, tem trazido vários desafios e oportunidades para a comunidade acadêmica e o mundo dos negócios, haja vista a necessidade de compreender, de forma automática, os sentimentos das pessoas a respeito de um produto, um serviço ou mesmo sobre pessoas ou fatos, para auxiliar no processo de tomada de decisão. Nos últimos anos, surgiram várias contribuições científicas para resolver problemas relacionados à análise de sentimentos. No entanto, poucas propostas consideram o fator espaço-temporal, isto é, a localização geográfica da fonte de informação ou da própria informação, bem como as possíveis mudanças de opinião ao longo do tempo. Os trabalhos que consideram o fator espacial tomam como base mensagens já geocodificadas, contudo, são poucas as fontes de informações que dispõem de mensagens georeferenciadas. Neste contexto, este trabalho propõe uma abordagem de análise de sentimentos que explora os fatores espaço-temporal para melhor sumarizar o sentimento detectado em uma grande quantidade de microtextos obtidos daWeb. A abordagem utiliza técnicas de Recuperação da Informação Geográfica (GIR) e técnicas de Análise de Sentimentos para detectar localizações geográficas e a polaridade dos sentimentos através de evidências textuais contidas nos microtextos, oferecendo mecanismos de visualização espacial do sentimento em diversas regiões geográficas. A análise espaço-temporal possibilita visualizar mudanças de sentimento ocorridas em diversas regiões geográficas ao longo do período analisado. / The dissemination of social communication means on the Web, such as blogs, discussion forums, product evaluation sites, microblogs and social networks, provides a never before seen volume of opinionative data in digital format. Not structured in its majority, this amount of data, has brought several challenges and opportunities for the academic community and the business world, considering the need for understanding, in an automatic form, people’s sentiments concerning a product, a service or even other people or facts, in order to facilitate the decision making process. In the recent years, several scientific contributions to solve sentiment analysis related problems were suggested. However, only a few of them consider the spatial-temporal factor, which is the geographical location of the information source or even of the information itself, as well as the possible opinion changes throughout time. The works that consider the spatial factor often assume the messages are already geocoded. However, it could be a problem, since only a few information sources provide georeferenced messages. In this context, this work proposes a sentiment analysis approach which explores the spatial-temporal factor in order to better summarize the sentiments detected in a great amount of microtexts obtained from the Web. The approach uses Geographic Information Retrieval (GIR) and Sentiment Analysis techniques for the detection of geographic locations and sentiment polarity through textual evidences contained in the microtexts. The spatialtemporal analysis enables the visualization of sentiment changes which occurred in several geographic regions throughout the analyzed time period.