JCoRe Token Boundary Detection Anaylsis Engine
A machine learning based token boundary detector
JTBD is a ML-based sentence splitter. It can be retrained on supported training material and is thus neither language nor domain dependent.
JTBD is based on a slightly modified version of the machine learning toolkit MALLET (Version 2.0.x). The necessary libraries are included in the executable JAR (see below) and accessible via the JULIE Nexus artifact manager.
To run JTBD just run the self-executing jar "jtbd-<version>.jar". This will show the available modes.
For further information please refer to the documentation, JTBD-x.pdf.