Interpreting intrinsic anology evaluation result

[seyoseyoseyo]

Hi, I'm having some trouble trying to interpret the output I get with my anology using BATS dataset. What does landing_a, landing_a_prime, landing_b, and landing_b_prime mean?

Also, the dataset only contain pairs of values like able->unable, so these are b and b prime? How does the program know what is a and a prime respectively?

Thanks a lot in advance.

[undertherain]

Hi this is additional statistics to help interpret "how model goes wrong" in original implementation Mykolov was excluding the words from the question from the possible answer. That is e.g. for king - man + woman , these words "king", "man" and "woman" can't be answers. If we don't do this, the resulting vector often lands into one of these source words and that's what is being counted. All pairs are combinatorially expanded into pairs of pairs