On many websites users can personally contribute information, ranging from short text messages to photos and videos. Users can see the information contributed by others and respond to it. These social media actively engage their community in the structuring of the collection by making use of collaborative annotation methods. Next to an improved description of the collection, collaborative annotations give insight in the personal preferences of individual users. Through all interactions with the data, users leave traces that can be exploited by the system to learn this preference and personalise social media access for each individual. This thesis contributes to the understanding of social media and collaborative annotation data by studying various data filtering tasks. Different types of collaborative annotations are used to adapt the collection access to the preference of individual users. The deployed data filtering methods are used as a means to learn about the factors that contribute to the accessibility of the information in the system. The results in this thesis show that small variations in data type, user interface and other system aspects appear to have large influence on the access possibilities of social media. By increasing the understanding of collaborative annotation data and the aspects that influence this data, this thesis has been able to improve existing data filtering methods and propose new opportunities for effective personalised access to social media.