Using social data as context for making recommendations (semantics of people and culture)
General Material Designation
[Thesis]
First Statement of Responsibility
Noor, Salma
Subsequent Statement of Responsibility
Martinez, Kirk
.PUBLICATION, DISTRIBUTION, ETC
Name of Publisher, Distributor, etc.
University of Southampton
Date of Publication, Distribution, etc.
2013
DISSERTATION (THESIS) NOTE
Dissertation or thesis details and type of degree
Ph.D.
Body granting the degree
University of Southampton
Text preceding or following the note
2013
SUMMARY OR ABSTRACT
Text of Note
This research explores the potential of utilising social Web data as a source of contextual information for searching and information retrieval tasks. While using a semantic and ontological approach to do so, it works towards a support system for providing adaptive and personalised recommendations for Cultural Heritage Resources. Most knowledge systems nowadays support an impressive amount of information and in case of Web based systems the size is ever growing. Among other difficulties faced by these systems is the problem of overwhelming the user with a vast amount of unrequired data, often referred to as information overload. The problem is elevated with the ever increasing issues of time constraint and extensive use of handheld devices. Use of context is a possible way out of this situation. To provide a more robust approach to context gathering we propose the use of Social Web technologies alongside the Semantic Web. As the social Web is used the most amongst today's Web users, it can provide better understanding about a user's interests and intentions. The proposed system gathers information about users from their social Web identities and enriches it with ontological knowledge and interlinks this mapped data with LOD resources online e.g., DBpedia. Thus, designing an interest model for the user can serve as a good source of contextual knowledge. This work bridges the gap between the user and search by analysing the virtual existence of a user and making interesting recommendations accordingly. i This work will open a way for the vast amount of structured data on Cultural Heritage to be exposed to the users of social networks, according to their tastes and likings.