Related Communities:

SemiMod Project

INTAS Grant IR-97-1109

Modelling and Management of Semi-Structured Data for Dynamic World-Wide-Web Applications (SEMIMOD)

Project starting date

  • March 1999

Project duration

  • 24 months

Project co-ordinator

  • Yannis Manolopoulos, University of Thessaloniki, Thessaloniki, Greece

Participating teams

  • University of Thessaloniki (Yannis Manolopoulos)
  • University of Rome (Paolo Atzeni)
  • St. Petersburg State University (Boris Novikov)
  • Institute for Problems of Informatics of the Russian Academy of Science (Leonid Kalinichenko)
  • Yerevan University (Manuk Manukyan)

Objectives

The field of semi-structured data is new, and dynamic aspects have not been examined extensively yet. Specifically,

  1. The existing data models, although extremely flexible, are also extremely terse.
  2. Consistency issues have not studied been at all.
  3. Temporal issues have not been incorporated in semi-structured data models.
  4. Efficient storage and indexing structures have not been developed and evaluated.

The proposed project addresses problems related to these issues. The objectives of the project are summarized as follows.

  1. To develop a canonical model for integration of heterogeneous sources (models) of semi-structured data and incorporation of temporal issues related to semi-structured data. To develop methods of describing existing semi-structured data in the framework of the canonical model.
  2. To develop methods of semantic identification and composition of pre-existing semi-structured components in course of the information systems development. To study consistency issues for heterogeneous distributed information systems (including Web-based) with semi-structured data and to develop techniques ensuring the consistency.
  3. To develop efficient dynamic indexing techniques for simple and temporal semi-structured data and materialised views. To study the performance of such techniques. To investigate and to develop techniques for wrapping and mediating semi-structured data.

The duration of the project is 24 months. During this period, five research teams (from the University of Rome, the Institute for Problems of Informatics of the Russian Academy of Science in Moscow, St. Petersburg State University, Yerevan University and University of Thessaloniki) will cooperate in completing four tasks, each one subdivided to a number of subtasks. Four meetings are scheduled during which the teams will discuss details of the progress of each task. Tasks 1 and 2 deal with conceptual issues of semi-structured data (in which Rome, Moscow and Yerevan have strong background), while tasks 3 and 4 deal with structuring and performance issues of the this kind of data (in which St. Petersburg and Thessaloniki have strong background). These two approaches should be combined in order to design and build powerful dynamic world-wide-web applications.

Task 1 is devoted to the development of the the canonical model for integration of heterogeneous models of semi-structured data. The participating teams are: Moscow, Rome, Yerevan.

Task 2 is devoted to the development of methods of semantic identification and composition of pre-existing semi-structured components in course of the information systems development. The participating teams are: Moscow, Rome, Yerevan.

Task 3 is devoted to the development of techniques ensuring the consistency of heterogeneous distributed information systems with semi-structured data. The participating teams are: St.Petersburg, Rome, Thessaloniki.

Task 4 is devoted to the investigation of performance issues of semi-structured data. The participating teams are: Thessaloniki, St.Petersburg, Rome.

The results of the project will appear in a number of technical reports and research papers. A prototype system incorporating some of the new ideas will also be developed.

Main Page

Main page of the project is located at the University of Thessaloniki http://skyblue.csd.auth.gr/INTAS

Publications of the Moscow group

Supported by Synthesis Group