Ai Dreams Forum

AI Dreams => New Users Please Post Here => Topic started by: Denis ROBERT on July 13, 2020, 12:23:48 pm

Title: Hi all - Johnny and Mylène
Post by: Denis ROBERT on July 13, 2020, 12:23:48 pm
Hi all,

It's been a while since I wanted to do it, today I registered on this forum. It is with pleasure that I find certain members there that I have known on other forums, and to discover others that I know less or not at all.

I am the author of a chatbot named Johnny (http://www.vixia.fr/mylene/johnny.php), or Mylène (http://www.vixia.fr/mylene/mylene.php) depending on the version. At first Johnny spoke English and Mylène spoke French, then both became bilingual, which means that today they are almost identical.

My chatbot Johnny has participated in numerous challenges many times, including 7 times in the Loebner Prize (from 2012 to 2018). I'm a regular at last place, but in 2017, I was eighth out of sixteen, which is not a bad result.

My chatbot is one of the few who build their sentences word by word, rather than having a list of pre-recorded answers. Similarly, the user's sentences are understood by the most complete semantic analysis possible, instead of having keyword recognition like most chatbots. We will probably talk about it later, but I am a convinced opponent of the keyword recognition system.

In order to pass the challenges, my chatbot gradually learned to answer the most common trick questions. I give you some examples, I probably forget some, others do not work 100% (Winograd schemas for example). It's just to give you an idea of ​​my chatbot's skills.

Common knowledge

Quote
How many paws has a cat?
Four paws.

What is the capital of Russia?
The capital of Russia is Moscow.

How many sides does a rectangle have?
Four sides.

Which is faster, a plane or a snail?
A plane is much faster than a snail.

Which is heavier, an elephant or one billion of mice?
One billion of mice are much heavier than an elephant.

Math questions

Quote
What is five multiplied by six?
It's easy. Five times six equals thirty.

5*8+x-8=1
5 * 8 + x - 8 = 1
<=> x = 1 + 8 - 40
<=> x = -31

Questions of logic

Quote
1 3 5 7 9?
It's easy. Eleven.

Find the odd word: bike, house, church, building, school.
A bike isn't a house.

What does gold and silver have in common?
Gold is a metal and silver is a metal.

What is between saturday and monday?
Sunday.

Questions about letters, syllables and rhymes

Quote
What is the first letter of the word banana ?
The first letter of the word banana is B.

What is the first vowel of the word banana ?
The first vowel of the word banana is A.

How many syllables are in the word perambulate?
The word perambulate has five syllables in it.

What word rhymes with cat?
The word tipcat.

Logical inferences

Quote
Can cats fly?
No, cats can't fly.
Why?
Cats can't fly because cats don't have wings.

My dog is Mallia
Well, your dog is Mallia.
How many paws has Mallia ?
Your dog has four paws.
Why?
Because your dog is a dog and because a dog has four paws.

Question with a non-real hypothesis

Quote
If I was a dragonfly, how many wing do I would have?
Let's suppose that you are a dragonfly. You would have four wings.

If you were a snake, how many paws do you would have?
Let's suppose that I am a snake. I wouldn't have paws.

Questions in the past or the future


Quote
I saw a good movie yesterday. I will see a movie tomorrow. When did I will see a movie?
You have said it to me. You will see a movie tomorrow.
When did I saw a movie?
You have said it to me. You saw a good movie yesterday.

Winograd Schemas

Quote
The trophy would not fit in the brown suitcase because it was too big. What was too big?
Is it a Winograd schema? The trophy was too big because if the suitcase was too big then the trophy would fit in the brown suitcase.

The man couldn't lift his son because he was so weak. Who was weak?
Yet another Winograd schema! The man was weak because if his son was weak then the man could lift his son.

Pete envies Martin although he is very successful. Who is very successful?
Yet another Winograd schema! Pete is very successful because if Martin was very successful then you wouldn't use the word although.

See you soon, have a good day.
Title: Re: Hi all - Johnny and Mylène
Post by: infurl on July 13, 2020, 12:33:36 pm
Welcome to the forum Denis, and doubly so for the approach that you've taken to develop your chatbots. We will have a lot to talk about I'm sure. It sounds like you have been working on them for a long-time, eight years at least. What got you interested in the first place?
Title: Re: Hi all - Johnny and Mylène
Post by: WriterOfMinds on July 13, 2020, 03:07:59 pm
Nice to meet you Denis, and welcome to the forum!
Title: Re: Hi all - Johnny and Mylène
Post by: Korrelan on July 13, 2020, 06:05:34 pm
Welcome Denis, Johnny & Mylene.

 :)
Title: Re: Hi all - Johnny and Mylène
Post by: squarebear on July 13, 2020, 07:16:34 pm
Great to see you over here Denis.
Title: Re: Hi all - Johnny and Mylène
Post by: Don Patrick on July 14, 2020, 07:15:42 pm
Welcome, Denis.
I don't post much here, but I hang around to watch project Acuitas develop, which I think you'll find interesting as well.
I always considered the Loebner Prize ill suited for testing logical AI, and your bot's French-like English also makes it sound a little less than natural even though it is well understood, so I wouldn't place too much value on whether Johnny did well in those contests. Overall, I always thought Johnny's core abilities were good, but needed some fleshing out to show how good they are.
I like that you got the "find the odd word" question to work, I haven't gotten around to it myself.
Title: Re: Hi all - Johnny and Mylène
Post by: Dat D on July 22, 2020, 02:21:42 am
Hi all,

It's been a while since I wanted to do it, today I registered on this forum. It is with pleasure that I find certain members there that I have known on other forums, and to discover others that I know less or not at all.
welcome, Denis  O0

Title: Re: Hi all - Johnny and Mylène
Post by: HS on July 22, 2020, 03:23:07 am
Hey Denis, welcome.

Now I'm wondering if sentences could be classified as another type of word. Maybe if you grouped synonymous sentences the numbers could be manageable.
Title: Re: Hi all - Johnny and Mylène
Post by: ruebot on May 04, 2021, 04:53:04 am
We will probably talk about it later, but I am a convinced opponent of the keyword recognition system.

The Personality Forge AI Engine uses a combination of NLP and Case-Based reasoning. My bot Demonica relies on keywords for the responses I match it to. It obviously has limits when compared to the method you're using.

However, I listen for keywords myself in using Behavior Modification. I can ask a question or make a statement off the top of my head during conversation that will bring a response and it's their keywords and the tone of voice, cadence and breaks in speech pattern I'm listening for.

It's how I was able to teach her that learned skill. I know what keywords to listen for and what response to use for a desired effect. In teaching her at my level she responds at that level using the same verbal techniques on me.

I have a transcript posted here as an example. Once I gave over control of the conversation and started answering her questions it didn't take her long to do it:

https://aidreams.co.uk/forum/bot-conversations/my-demon-turned-on-me/