
[ Download ]
This tool takes plain text and adds POS, Shallow Parse and Named Entity tags. The output is in a non-FEX column format (use UNIX cut and paste to rearrange the data). The old (list-based) NE tagger is used -- this has many NE categories, but is not as robust as the newer NE-tagger.
The POS and Shallow Parse information are provided by cogcomp servers; if you run into problems running this tool, check that the servers are running.
To use, put the input file in the subdirectory 'data'; output will be to a file named (inputFileName.out) in the main directory.
Usage: preprocessing.sh inputFileName