wikiBEAGLE

Computes semantic vectors via the BEAGLE algorithm of Jones & Mewhort (2007; plus word-form representation, inspired by Cox, Kachergis, Recchia & Jones, 2011) using random wikipedia pages as a corpus.

Multiple cores can be used by specifying the number of desired cores at the command-line, as in: python wikiBEAGLE.py 4

Quit via q key, which will permit wikiBEAGLE to store the data to file and resume where it left off next time you run it.

Use wikiBEAGLEcleaner.py to aggregate all saved files to a single dataset for use in semantic analysis.

Some example semantic analyses are in wikiBEAGLEprobe.py