The future versions of the system will:

• rely on an open source XML database management system
• include a GUI
• operate over text windows as opposed to sentences
• improve the similarity measure with acronym recognition and LSI
• classify term relations in addition to terms