Good idea Vince.  That may be especially useful
for newly created chatbots that need to practice
to do better in the official chatbot contests.

General comment:
(Drats! Spelled “advertisments” wrong above.)

When I was working on creating a “replacement” contest for the CBC, I purchased some league/ladder software for use with the competition. Sadly, things got put on hold due to family issues at the time, but I still have the software, and still plan to make use of it. I’ve got too many big projects going on right now to do something like this by myself, but I’m certainly willing to contribute what I can to support Vince’s idea. Anyone else want to help out?

Robo Chat Challenge Website Updates:

Super 10 Results have been updated on the website which featured in the Final Stage of Robo Chat Challenge and the Audience Poll is now closed for general public or the bot masters.

http://www.robochatchallenge.com/results.php

Regards,

Robo Chat Challenge Team

Thanks for the results Haris. Are we able to see the transcripts of the various bots online at all?

It is truly an honor to make it to the final round with the top ten chatbot masters in the world.  I would like to acknowledge the other Super chatbot masters 11 through 21 as well and thank them all for playing.  I hope to see you all in this contest again next year.

Follow us on our Twitter and Facebook pages:

http://www.twitter.com/RoboChatContest

http://www.facebook.com/RoboChatChallenge

Robo Chat Challenge Team

Steve Worswick - Jan 23, 2013:

Thanks for the results Haris. Are we able to see the transcripts of the various bots online at all?

Dear Steve,

At this point of time, it is not possible to share the transcripts but we are looking into it and how do you found the results? Just a mere question!!

Regards,

Robo Chat Challenge Team

The transcripts would be most interesting as I see one person here mentioned there was no evidence that a judge spoke to their bot and I know two other contestants reporting a similar story.

Roger Davie - Jan 24, 2013:

The transcripts would be most interesting as I see one person here mentioned there was no evidence that a judge spoke to their bot and I know two other contestants reporting a similar story.

First off congratulations to the finalists whatever the means by which they were selected. Certainly the finalists are all great examples of the genre and the technology and each and every one deserves to be there no doubt. That having been said, I agree that there was no indication that anyone ever visited RICH that might have been a judge, given the contest guidelines as a template for the types of questions that might have been asked.

Vince

NOTE: The site appears to be gone \ down

Please note that the Competition Results have nothing to do with the Audience Poll results.

Regards,

Robo Chat Challenge Team

Haris Lodhi - Jan 24, 2013:

.. how do you found the results?

Naturally, I am pleased to be ranked number one but also a little surprised, as I too can see no visits to Mitsuku from anyone asking competition type questions.

The only visit I could see from anyone resembling a judge was someone from the United Arab Emirates who asked, “If happiness was the national currency, what work would you do that would make you rich?” and later came back to ask something along the lines of , “Can a cat do fly?”. Were these part of the contest questions?

One of the reasons people enter these contests is so they can see in what areas their bots need improving and certainly if I had been placed in postitions 11 through 21, I would like to see where my chatbot needed more work. In fact, unless there were exactly 22 questions and Mitsuku scored 100 in each one, I would like to see where she was successful and where she failed.

I notice http://www.robochatchallenge.com/ has been down for about a day now. Everything ok?

Steve Worswick - Jan 25, 2013:

Haris Lodhi - Jan 24, 2013:

.. how do you found the results?

Naturally, I am pleased to be ranked number one but also a little surprised, as I too can see no visits to Mitsuku from anyone asking competition type questions.

The only visit I could see from anyone resembling a judge was someone from the United Arab Emirates who asked, “If happiness was the national currency, what work would you do that would make you rich?” and later came back to ask something along the lines of , “Can a cat do fly?”. Were these part of the contest questions?

One of the reasons people enter these contests is so they can see in what areas their bots need improving and certainly if I had been placed in postitions 11 through 21, I would like to see where my chatbot needed more work. In fact, unless there were exactly 22 questions and Mitsuku scored 100 in each one, I would like to see where she was successful and where she failed.

I notice http://www.robochatchallenge.com/ has been down for about a day now. Everything ok?

Sorry for the inconvenience, as Robo Chat Challenge is down for maintenance. We will be back shortly. In case you have any queries, you can contact on this thread or email (if it’s deemed necessary).

Regards,

Robo Chat Challenge Team

Haris Lodhi - Jan 25, 2013:

Steve Worswick - Jan 25, 2013:

Haris Lodhi - Jan 24, 2013:

In fact, unless there were exactly 22 questions and Mitsuku scored 100 in each one, I would like to see where she was successful and where she failed.
ok?

I was slightly curious about the math myself. If I remember correctly, points awarded were in the 1 to 5 range.

Vince

If I remember correctly, it was 100 points for a spot on answer, 75 points for a near miss, 50 points for something fairly reasonable, 25 points for a hint that the chatbot had at least understood and 0 points for a nonsense answer.

I think 25 points was also awarded for “I don’t know” type answers which to me would open the floodgates for a bot to give these kind of answers for whatever was asked of it in the hope it would make the cut.

Steve Worswick - Jan 25, 2013:

If I remember correctly, it was 100 points for a spot on answer, 75 points for a near miss, 50 points for something fairly reasonable, 25 points for a hint that the chatbot had at least understood and 0 points for a nonsense answer.

I think 25 points was also awarded for “I don’t know” type answers which to me would open the floodgates for a bot to give these kind of answers for whatever was asked of it in the hope it would make the cut.

Yes, Steve you are correct but to some extent as 25 Marks were not to be given for a simple answer “I don’t know” but the relevancy to the question should also be maintained.

Regards,

Robo Chat Challenge Team