Visualising results in a tabular format

The .score files produced by TRACER can be transformed into a more human-readable format with a script that prints the corresponding text segment next to the reuse ID, as well as the number of overlapping features and the percentage of similarity. In the terminal, navigate to the TRACER main folder and type:

java -cp tracer.jar eu.etrap.tracer.postprocessing.DefaultOutputterMain /.../filename.txt /.../filename.score

Where:

• /.../filename.txt is the path to the main text file;
• /.../filename.score is the score file.

The generated file, accessible from /data/corpora/myfolder/, has .expanded as its suffix (i.e., filename.score.expanded) and contains four columns separated by tabs:

RUID 1, RUID 2, FEATUREOVERLAP, SIMTHRESHOLD, TEXT 1, TEXT 2

This .txt output can be duplicated and imported into a .csv file for the gold standard analysis