That post in turn led to an experiments link on the CLL site, where you can participate in experiments (typical completion time: 5 minutes).
That last post describes Latent Semantic Analysis, a set of algorithms assessing the similarity of words to other words. I especially liked this conclusion:
After training, LSA performed at 64.4% correct on a multiple choice test of synonymity taken from TOEFL (in contrast, humans score around 64.5% on average on this test, which is frequently used as a college entrance examination of English proficiency in non-native speakers. By this metric, LSA would be admitted to many major universities!)