The Text Categorization tool 2009 version is the 3rd offical public rlease.It was developed in pure Java, capable of handling UTF-8. Belows are some specifications of this tool.

System

• Upgrade to Java 1.6.0.13
• Upgrade to HSqlDb 1.8.0.10
• Provides scripts for command line tools

Data

• Used MEDLINE.2009 for citations created in years of 2006, 2007, 2008
• Used Metathesaurus.2008AB
• Used lsi2009.xml
• Used new designed algorithm to generate optimum stDocuments
• Used the latest data set for JDI, STI, and STRI
• Updated the default value of Mac. normalized count
• Provides completed data set of tcData.2007 and tcData.2008 to run on TC.2009

Java APIs/Tools/Classes

• New Tool StWsd
• New Java APIs of StWsd

Tool Options

• New options to run specified data set, TC.2007 or TC.2008 in Jdi, Sti, Stri, and StWsd
• New options of specifying score types in StWsd
• New Options of specifying STI or STRI in StWsd
• New Options of using ambiguous sentences in StWsd
• New Options of show details of ST scores in StWsd