The 2nd experiment is to compare the translation efficacy of the machine and the distributed human intelligence. This is a meta experiment that pitches the signal analysis of the utterances of the geese (and human attempts)--that is, the traditional animal linguistic methodology/AI methodology--against the interpretations of the distributed network of lay participants. The metrics for comparing the machine and distributed human interpretation will rest on consistency measures. This is naturally weighted strongly to the advantage of human based interpretation simply because humans can effortlessly exploit the video and therefore spatial, proximity and deitic information that is systematically under represented in signal databases. Again, this could potentially be an important result that would challenge the computational linguistics community, and the political realms over which they have inordinate influence (such as national security and intelligence communities primarily). There is no other translation experiment that compares these two methods on the same data in an ongoing manner.

 

 

Previous Slide :: Next Slide

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32