Authors:
Andrea Romei
and
Franco Turini
Affiliation:
University of Pisa, Italy
Keyword(s):
Data mining, Knowledge discovery, Inductive databases, XML, XQuery, Query language.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Data Mining
;
Databases and Information Systems Integration
;
Enterprise Information Systems
;
Knowledge Representation and Reasoning
;
Sensor Networks
;
Signal Processing
;
Soft Computing
;
Symbolic Systems
Abstract:
The rapid growth of semi-structured sources raises the need of designing and implementing environments for knowledge discovery out of XML data. This paper presents an Inductive Database System in which raw data, mining models and domain knowledge are represented as XML documents, stored inside XML native databases. In particular, we discuss our experiences in the design and development of XQuake, a mining query language that extends XQuery. Features of the language are an intuitive syntax, a good expressiveness and the capability of dealing uniformly with raw data, induced and background knowledge. The language is presented by means of examples and a sketch of its implementations and the evaluation of its performance is given.