AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

The finalists?
 
 

Wasn’t today the day we heard who the finalists were? Any ideas anyone?

 

 
  [ # 1 ]

Results were posted at http://xa.yimg.com/kq/groups/7477452/1679993586/name/Loebner Prize 2011 Selection Results.docx

Rosette, Zoe, ChipVivant, and Tutor are the names of the 4 bots going to the final round. Full scores:

Scores in 2nd round based on audience voting:
Rosette 18.01%
Zoe 15.52%
ChipVivant 12.98%
Tutor 12.30%
Mitsuku 11.84%
Cleverbot 10.57%
Alice 6.96%
Eugene_Goostman 6.95%
UltraHal 4.88%

Scores of bots who didn’t progress to 2nd round based on 2010 scoring system:
8pla 5.5
Trane 5.5
Adam-L 3.5
SEARS 0

 

 
  [ # 2 ]

8pla.net (8pla) scored high enough, yet did not progress to the 2nd round, to break a tie?

Rosette 18.01%
Zoe 15.52%
ChipVivant 12.98%
Tutor 12.30%
Mitsuku 11.84%
Cleverbot 10.57%
Alice 6.96%
Eugene_Goostman 6.95%
8pla 5.5
Trane 5.5
UltraHal 4.88
%
Adam-L 3.5
SEARS 0 

8pla.net confused the Loebner Prize judges so much, it practically passed the Turing Test.

 

 

 
  [ # 3 ]

It’s a bit confusing, but from what I understand the first round was point-based: 1 point for good answers, 0.5 points for border-line answers, 0 for bad answers. The second round was based on an audience voting and came out with a percentage score. So 5.5 is the first round score and 4.8% is a second round score. What we’re missing is the first round score for all the other bots.

 

 
  [ # 4 ]

@Robert:

A score of 5.5 is a round 2 score, since it is in the top 10 scores of round 1 of the contest.

Only 9 progressed instead of ten, because, “responses” by 8pla.net, “could not be separated” by the human judges at the University of Exeter.

Awesome!  8pla.net passed the Turing Test… Mission Accomplished!

What I don’t understand is ....
How are “technical limitations of the voting technology” the fault of the contestant?

 

 
  [ # 5 ]

Nice to see at least 3 out of the top 5 were AIML bots smile

 

 
  [ # 6 ]
8PLA • NET - Jul 5, 2011:

@Robert:

A score of 5.5 is a round 2 score, since it is in the top 10 scores of round 1 of the contest.

Only 9 progressed instead of ten, because, “responses” by 8pla.net, “could not be separated” by the human judges at the University of Exeter.

Awesome!  8pla.net passed the Turing Test… Mission Accomplished!

What I don’t understand is ....
How are “technical limitations of the voting technology” the fault of the contestant?

No, your bot didn’t make it past round one. Not quite sure why you think it passed the Turing Test? Your bot and Trane could not be separated (presumably because they were AIML clones?) not your bot and a human.

 

 
  login or register to react
‹‹ Is There An Entrants Page      Loebner Survey ››