Semantically annotating the personal desktop

Abstract

The advent of the World-Wide Web brought with it a proliferation of information from e-mail, forums, chat, sites that rapidly led to information overload and a subsequent storage and maintenance problem on users personal computers. The desktop has become a repository of data that hosts various types of files. The recent massive increase in data has resulted in a continuous attempt to enhance our data organisation techniques and hence to the development of personal information management software. In this paper we present an overview of data organisation techniques related to personal data management that have been an active research area for decades. We will look at how personal information managers handle different types of files, and abstract these file types into a single user interface. Despite their advanced user interfaces, we argue that traditional personal information managers tend to be very domain specific and lack in user adaptability. To address these limitations we propose a semantic desktop system that exploits the flexibility of semantic web technologies, introduces the concept of a Personal Ontology to aid in data organisation and can be used by other desktop applications such as information retrieval and intelligent software agents.