Bagaimana memilih satu Sistem Data Mining?

• Commercial data mining systems have little in common

· Different data mining functionality or methodology

· May even work with completely different kinds of data sets

• Need multiple dimensional view in selection
• Data types: relational, transactional, text, time sequence, spatial?
• System issues

· running on only one or on several operating systems?

· a client/server architecture?

· Provide Web-based interfaces and allow XML data as input and/or output?

· Data sources

· ASCII text files, multiple relational data sources

· support ODBC connections (OLE DB, JDBC)?

· Data mining functions and methodologies

· One vs. multiple data mining functions

· One vs. variety of methods per function

· More data mining functions and methods per function provide the user with greater flexibility and analysis power

· Coupling with DB and/or data warehouse systems

· Four forms of coupling: no coupling, loose coupling, semitight coupling, and tight coupling

· Ideally, a data mining system should be tightly coupled with a database system

· Scalability

· Row (or database size) scalability

· Column (or dimension) scalability

· Curse of dimensionality: it is much more challenging to make a system column scalable that row scalable

· Visualization tools

· “A picture is worth a thousand words”

· Visualization categories: data visualization, mining result visualization, mining process visualization, and visual data mining

• Data mining query language and graphical user interface

· Easy-to-use and high-quality graphical user interface

· Essential for user-guided, highly interactive data mining