The advent of the World Wide Web (WWW) has meant there is a growing need to link information held in databases and files with information held in other types of structure [2]. This poster presents a toolkit which enables information held in tables within documents to be linked with data held in conventional databases. The toolkit addresses the problems of:- •
extracting tabular data from various types of document e.g. semi-structured text [4], HTML and spreadsheets
•
providing a standard interface to these tables via a standard query language
•
extracting useful table meta-data from the table and accompanying text.