Supporting Personalised Information Exploration through Subjective Expert-created Semantic Attributes

Abstract

Ordinary users are finding it increasingly difficult to explore the large volumes of diverse data they encounter in their everyday lives. Techniques based on data mining algorithms are useful but they tend to be too complex for casual users to work with effectively. Furthermore, these techniques don?t allow the user to engage with the information using semantics meaningful to them. Semantically enriched and personalized data exploration is seen as an essential step to support such users. Moreover, by allowing these users to leverage and personalize the subjective insights and knowledge of experts, more relevant and useful information can be discovered and interesting correlations drawn. In order to support these domain specific explorations, a prototype architecture named SARA (Semantic Attribute Reconciliation Architecture) has been built, and its underlying methodology, implementation and initial evaluation are described within this paper.