iDM: a unified and versatile data model for personal dataspace management

  • Authors:

  • Jens-Peter Dittrich;Marcos Antonio Vaz Salles

  • Affiliations:

  • ETH Zurich, Zurich, Switzerland;ETH Zurich, Zurich, Switzerland

  • Venue:
  • VLDB '06 Proceedings of the 32nd international conference on Very large data bases
  • Year:
  • 2006

Quantified Score

Hi-index 0.02

Visualization

Abstract

Personal Information Management Systems require a powerful and versatile data model that is able to represent a highly heterogeneous mix of data such as relational data, XML, file content, folder hierarchies, emails and email attachments, data streams, RSS feeds and dynamically computed documents, e.g. ActiveXML [3]. Interestingly, until now no approach was proposed that is able to represent all of the above data in a single, powerful yet simple data model. This paper fills this gap. We present the iMeMex Data Model (iDM) for personal information management. iDM is able to represent unstructured, semi-structured and structured data inside a single model. Moreover, iDM is powerful enough to represent graph-structured data, intensional data as well as infinite data streams. Further, our model enables to represent the structural information available inside files. As a consequence, the artifical boundary between inside and outside a file is removed to enable a new class of queries. As iDM allows the representation of the whole personal dataspace [20] of a user in a single model, it is the foundation of the iMeMex Personal Dataspace Management System (PDSMS) [16, 14, 47]. This paper also presents results of an evaluation of an initial iDM implementation in iMeMex that show that iDM can be efficiently supported in a real PDSMS.