Enhancing E-learning platforms with social networks mining

Social Networks appeared as an Internet application that offers several tools to create a personal virtual profile, add other users as friends, and interact with them through messages. These networks quickly evolved and won particular importance in people lives. Now, everyday, people use social networks to share news, interests, and discuss topics that in some way are important to them. Together with social networks, e-learning platforms and related technologies have evolved in the recent years. Both platforms and technologies (social networks and e-learning) enable access to specific information and are able to redirect specific content to an individual person. This dissertation is motivated on social networks data mining over e-learning platforms. It considers the following four social networks: Facebook, Twitter, Google Plus, and Delicious. In order to acquire, analyze, and make a correct and precise implementation of data, two different approaches were followed: enhancement of a current e-learning platform and improvement of search engines. The first approach proposes and elaborates a recommendation tool for Web documents using, as main criterion, social information to support a custom Learning Management System (LMS). In order to create the proposed system, three distinct applications (the Crawler, the SocialRank, and the Recommender) were proposed. Such data will be then incorporated into an LMS system, such as the Personal Learning Environment Box (PLEBOX). PLEBOX is a custom platform based on operating systems layout, and also, provides a software development kit (SDK), a group of tools, to create and manage modules. The results of recommendation tool about ten course units are presented. The second part presents an approach to improve a search engine based on social networks content. Subsequently, a depth analysis to justify the