Architectural approach for handling semi‐structured data in a user‐centred working environment

Today the amount of all kinds of digital data (e.g. documents and e‐mails), existing on every user's computer, is continuously growing. Users are faced with huge difficulties when it comes to handling the existing data pool and finding specific information, respectively. This paper aims to discover new ways of searching and finding semi‐structured data by integrating semantic metadata.

The proposed architecture allows cross‐border searches spanning various applications and operating system activities (e.g. file access and network traffic) and improves the human working process by offering context‐specific, automatically generated links that are created using ontologies.

The proposed semantic enrichment of automated gathered data is a useful approach to reflect the human way of thinking, which is accomplished by remembering relations rather than keywords or tags. The proposed architecture supports the goals of supporting the human working process by managing and enriching personal data, e.g. by providing a database model which supports the semantic storage idea through a generic and flexible structure or the modular structure and composition of data collectors.

Available programs to manage personal data usually offer searches either via keywords or full text search. Each of these existing search methodologies has its shortcomings and, apart from that, people tend to forget names of specific objects. It is often easier to remember the context of a situation in which, for example, a file was created or a web site was visited. By proposing this architectural approach for handling semi‐structured data, it is possible to offer a sophisticated and more applicable search mechanism regarding the way of human thinking.