Marjut was – her being the test addict she is – astounded at the Gender genie’s accuracy in determining her gender. Well, I decided to give it a shot and fed the longest text I could find into the machine.

The results: Words: 990, Score: 917. The Gender Genie thinks the author of this passage is female!

Analysis
Male Keywords Female Keywords
[the] 57 × +17 = 969 [with] 7 × -14 = -98
[a] 16 × +6 = 96 [’s] 6 × -5 = -30
[some] 5 × +6 = 30 [possessive pronoun] 3 × -3 = -9
[number] 7 × +5 = 35 [for] 7 x- 4 = -28
[it] 12 × +2 = 24 [not] or [n’t] 18 × -4 = -72

What can I say? According to Koppel and Argamon, the algorithm should predict the gender of the author approximately 80% of the time, says the machine. Which is sorta funny when contrasted with the data just below that statement (49.83% right, 50.17% unright). But you know the deal with best laid plans, don’t you?

Samantapaisia: