The British National Conference on Databases (BNCOD) was established in 1980 as a forum for research into the theory and practice of databases. This aim remains highly relevant today as database technology must meet the challenges of global applications that need to access and analyse data stored in multiple heterogeneous, autonomous - formation sources on the Web. th This volume contains the proceedings of the 19 BNCOD, held at the University of Shef?eld in July 2002. The conference theme this year is "Exploring the roles of database technology within the global computing infrastructure". There are 10 fu- length papers, 9 poster papers, and two invited talks. As in previous BNCOD conf- ences, there are a signi?cant number of papers from abroad, including authors from Brazil, France, Italy, The Netherlands, Spain, Sweden, and the USA. Our?rstinvitedspeakerisJiawei Hanwhois currentlyattheUniversityofIllinoisat Urbana-Champaign.Jiawei Han is a distinguishedresearcherin the area of data mining, and his invited talk addresses in particular data mining techniques for stream data.The motivation for this research is the vast amounts of data producedat high speed by many real-time systems, which for practical reasons cannot be stored on disk for off-line analysis. Thus, techniques are needed for real-time mining of such stream data. Jiawei Han'stalkaddressesrecentdevelopmentsinthis area,andthemajorresearchchallenges arising.

The British National Conference on Databases (BNCOD) was established in 1980 as a forum for research into the theory and practice of databases. This aim remains highly relevant today as database technology must meet the challenges of global applications that need to access and analyse data stored in multiple heterogeneous, autonomous - formation sources on the Web. th This volume contains the proceedings of the 19 BNCOD, held at the University of Shef?eld in July 2002. The conference theme this year is "Exploring the roles of database technology within the global computing infrastructure". There are 10 fu- length papers, 9 poster papers, and two invited talks. As in previous BNCOD conf- ences, there are a signi?cant number of papers from abroad, including authors from Brazil, France, Italy, The Netherlands, Spain, Sweden, and the USA. Our?rstinvitedspeakerisJiawei Hanwhois currentlyattheUniversityofIllinoisat Urbana-Champaign.Jiawei Han is a distinguishedresearcherin the area of data mining, and his invited talk addresses in particular data mining techniques for stream data.The motivation for this research is the vast amounts of data producedat high speed by many real-time systems, which for practical reasons cannot be stored on disk for off-line analysis. Thus, techniques are needed for real-time mining of such stream data. Jiawei Han'stalkaddressesrecentdevelopmentsinthis area,andthemajorresearchchallenges arising.

The British National Conference on Databases (BNCOD) was established in 1980 as a forum for research into the theory and practice of databases. This aim remains highly relevant today as database technology must meet the challenges of global applications that need to access and analyse data stored in multiple heterogeneous, autonomous - formation sources on the Web. th This volume contains the proceedings of the 19 BNCOD, held at the University of Shef?eld in July 2002. The conference theme this year is "Exploring the roles of database technology within the global computing infrastructure". There are 10 fu- length papers, 9 poster papers, and two invited talks. As in previous BNCOD conf- ences, there are a signi?cant number of papers from abroad, including authors from Brazil, France, Italy, The Netherlands, Spain, Sweden, and the USA. Our?rstinvitedspeakerisJiawei Hanwhois currentlyattheUniversityofIllinoisat Urbana-Champaign.Jiawei Han is a distinguishedresearcherin the area of data mining, and his invited talk addresses in particular data mining techniques for stream data.The motivation for this research is the vast amounts of data producedat high speed by many real-time systems, which for practical reasons cannot be stored on disk for off-line analysis. Thus, techniques are needed for real-time mining of such stream data. Jiawei Han'stalkaddressesrecentdevelopmentsinthis area,andthemajorresearchchallenges arising.