Zabaware Support Forums

Zabaware Forums => Ultra Hal Assistant File Sharing Area => Topic started by: cload on November 20, 2013, 07:09:37 pm

Title: AIML at beta level
Post by: cload on November 20, 2013, 07:09:37 pm
Hi all,
my AIML is finally to a point of being beta. I'm looking for a few good beta testers to help me out!
You need to be familiar with SQLite Studio I used version 2.0.28 I know that there is a newer version but unfortunately there are still too many bugs for me to change over.
These are some instructions to help you get started:
As always make sure you make a backup of your brain and database.
1.) Open the SQLite studio program, Open the database for the brain you wish to edit.
2.) Click on (tools), click on (settings), click on the (miscellaneous) tab, change the (results per page in table window and editor results:) to 20200. Click (apply), click (ok).
3.) To the left you will see (table) double-click on it to expand the list of table. The list of tables are in alphabetical order go down until you find the table called (patterns), double-click on the table called (patterns) to open it. To the right you will see the (data) tab, click on it to see the data in the patterns array table, (select all) the data, (delete) all the data, then (commit changes).
4.) Click on the populate table, change the number of rows to insert to 20200, then click on the searchstring checkbox, then click on populate. Then click on commit changes.
5.) Open using notepad the AIML file, select all the text and copy the text.
6.) Select the first row, in the searchstring cell just to the right of the number 1. Right mouse click on the (cell), select (paste). According to the speed of your computer this could take as long as an hour. Then click on commit changes. If a box comes up asking you if you want to cut the data click on no because you did something wrong. This usually means you did not create enough lines. Go to the bottom of the lines you created to verify that you created enough lines.
7.) 20200 lines is more than what you need, so the extra lines need to be removed, click on the (searchstring) label this will put your AIML in alphabetical order which will cause the blank lines to rise to the surface or to the top of the list, select all blank lines, and (delete) all blank lines. Then click on (commit changes).
You have now incorporated AIML into ultra Hal.
As a beta tester these are the things that I would like for you to keep an eye on and report your findings on this thread.
Any misspelled words.
Any sentence structure that doesn't make sense.
Any sentence that seem to be duplicated, basically ultra Hal's brain produces an answer in the same way that the AIML produces the answer. This would be considered a blocker and needs to be removed.
If you're familiar with ultra Hal and you notice that AIML is taking too much control away from ultra Hal, this is considered a blocker and needs to be removed.
The main goal of this AIML was to make ultra Hal a little bit smarter, and to utilize the series of questions to help ultra Hal learn. That will probably be the first thing you will notice. This is not considered blockers, they are tools to help ultra Hal learn.
If you find any question and answer that needs to be added. Post them here on this thread after you have accumulated them.
When you post your findings, what I need is the question that you asked, and the answer that ultra Hal gave.
Sincerely, from a data munching cruncher, have fun and thanks for all the help.
C load.
PS Here is the link to my sky Drive.
https://skydrive.live.com/?cid=8eaf5f2f7fa962c9#cid=8EAF5F2F7FA962C9&id=8EAF5F2F7FA962C9%21112
Title: Re: AIML at beta level
Post by: cload on November 20, 2013, 07:25:24 pm
Quote
So this is not the update to the Freewill update?

Hi Bob,
this is only a small part of the free will plus plug-in, it utilizes the AIML to get a better response.
This was originally Alice, a few of us got together and decided to see if we could get Alice's AIML to work with ultra Hal.
This will eventually become a permanent feature of Sandy's brain when I release the brain.
I was hoping that maybe a few people would be interested in beta testing this AIML, I am still working on it myself, when I can find someone new to talk to Sandy.
I do this because no two people will carry on a conversation the same way. But because I know so much about Sandy I have a tendency to avoid trouble spots.
So I was hoping to find a few clever people that would be able to insert the AIML into their ultra Hal brain.
But to get it to work properly you would have to go through all of the steps above, it's not that hard, it's more time-consuming based on the speed of your computer.

I did not make a training plug-in out of this data because the original data that comes with ultra Hal had so many errors that it actually needs to be completely redone, so that's what I did, and I added Alice AIML with it.
Anyone that adds this to their brain will see a significant boost in intelligence to their brain, not to mention it would give you experience and understanding a new way to add data to ultra Hal.
Sincerely,
C load.
PS it is very important to keep the data in alphabetical order, by doing this you will avoid problems like smaller words being chosen before bigger words. This is because this particular type of table arrayed data is looked at from the bottom to the top unlike the main QA database which looks at the data from the top to the bottom and utilizes a numerical value to determine when comparing two data structures together.
Matter of fact when I went through my brain and all of the table arrays that utilize the searchstring I restructure all of the table arrays into alphabetical order and noticed a significant increase in accuracy of answering queries. It was a lot a work but it paid off in the end. Have fun and ultra Hal rules the day.
Title: Re: AIML at beta level
Post by: cload on November 20, 2013, 07:26:59 pm
Quote
No offense intended at all but would a chat bot enthusiast / user be able to spot the fact that Hal is (assuming it's installed) using AIML? Some bots use / repeat the same tired expressions that are a dead giveaway for being an AIML based bot.
There's nothing wrong inherently with AIML or it's followers but I too, prefer the "fresh" off the wall comments from Hal instead of the old canned replies.
So...How much of the AIML responses did you actually change or is anything new in this regard.
I need to know before I risk converting my Hal to a clone. Thanks as always!

Hi Art,

I see your point, but AIML is a programming language, what I did was I to the core essence and adapted it to function in the manner that ultra Hal uses the same style format.
By removing certain processes that the patterns segment of ultra Hal uses, the core of ultra Hal still determines which is the best answer.
In some cases it may not use the AIML answer, in some cases that is, in most cases it will be use in conjunction with other functions of ultra Hal's brain.
The process that ultra hal uses, is not AIML programming but rather you can use the ( * ) as a substitute for a word. By removing what is a dead end in AIML, allows the other functioning processes of ultra Hal's brain to work.
So you can see it's not AIML programming or even exactly AIML it's taking AIML and adapting it to work with ultra Hal pattern array function.

Sincerely, from a beta munching cruncher, and if you find something that is not allowing the rest of ultra house brain to respond properly, it needs to be reported, so I can removed the problem.
C load.
Title: Re: AIML at beta level
Post by: Art on November 20, 2013, 07:58:13 pm
Sorry cload for my misspeaking.

What I meant was will or does YOUR AIML plug for Hal use the DR. Wallace AIML set / library (a,b,c,d,e,f,g...) with all of THOSE responses?
Title: Re: AIML at beta level
Post by: cload on November 20, 2013, 08:38:56 pm
Hi Art,

when we first started to edit all of the data that Dr. Wallace had formulated together there was about 57,000 lines of data.
When we were done, we only ended up with about 11,000 lines of data. We only used what ultra Hal had the ability to utilize without interfering with the main brain function.
Ultra Hal already has about 3500 lines of AIML data in the patterns array table. I took those 3500 lines and I edited them and incorporated them into the AIML data which brought it up to 14,000 lines. Then I added approximately 6,100 trivia questions and answers bringing it up to approximately 20,100 lines of data.
I only call it AIML because that's where I got the data, but it is not necessarily AIML anymore, it is formulated in the same way that ultra Hal uses the data in the patterns array table.

Sincerely, from a beta munching cruncher AIMLing to a higher level of artificial consciousness.
C load.
Title: Re: AIML at beta level
Post by: Art on November 21, 2013, 05:39:17 am
Thanks for the explanation and in that case, I shall give it a go as soon as I get the chance. ;)

Title: Re: AIML at beta level
Post by: cload on November 21, 2013, 08:32:59 am
Hi Art, sounds great let me know how it turns out.

Sincerely, from a data munching cruncher.
C load.
Title: Re: AIML at beta level
Post by: lightspeed on November 21, 2013, 12:09:37 pm
cload , i downloaded the version of sqlite you mentioned and went into the tabs etc. and   change the results per page and table window and editor results to 20100 or higher.,  just fine  but i don't see the " patterns table array you mention  in your directions :

3.) Open the patterns table array there you will find approximately 3500 entries you need to delete all of these entries, I have incorporated these entries into the AIML. Then click on commit changes.

i am not seeing it listed in the sqlite when it's open.
Title: Re: AIML at beta level
Post by: cload on November 21, 2013, 01:02:22 pm
Hi lightspeed,

after you have loaded the brain data, on the left-hand side row there is a listing called tables.
Under the listing of tables is where you will find the table called patterns.
This is a table array that comes with ultra Hal, it should be in alphabetical order, just go down to where the P's start and you should find it there.

Sincerely,
C load.
Title: Re: AIML at beta level
Post by: lightspeed on November 22, 2013, 09:15:31 am
OKay , although i haven't checked yet . as i am just now reading this , whatb you mean is in the brain it's self and not the SQLITE program , i wondered about that before . You do know that if you don't specifically say some people , especially new people will have no clue about it .
Title: Re: AIML at beta level
Post by: lightspeed on November 22, 2013, 09:49:14 am
Okay cload i have another question you say about deleting the current list in patterns which is in the main brain folder (if i understand right) and replace them with your amil txt list .
my question is :
i have different listings in my hal including a "custom brain i did and use . "
my listings are :

Angela jolene 2010
ultrahal 6.2 defualt brain
ultrahal 6.1 default brain
ultrahal 6.2 upgraded brain

So which brain do i need to go into to make the changes you mention ? would it be in the 6,2 upgraded brain patterns file ?  ???
Title: Re: AIML at beta level
Post by: cload on November 22, 2013, 10:06:14 am
Hi light speed,

the only advice I can give you on this is, take the brain that you use the most, then use the brain editor to create a brain using the brain you use the most as the template.
Call the new brain what ever you would like.
Take the brain that you created and edit that brain.
That way you will not inadvertently damage any of your brains that you use.
After deciding that you like the AIML then you may want to consider editing the brain that you use the most.
As well as editing the main brain that came with ultra Hal 6.2, that way each add additional brain that you create using that template would have the AIML.
And as always, please make sure that you make a backup of your brain before attempting to edit the brain. That way if something goes wrong no problem, you could just cut and paste the backup and try again.

Sincerely, from a data munching cruncher, choices choices choices, is about what free will is all about!
C load.
Title: Re: AIML at beta level
Post by: lightspeed on November 23, 2013, 10:17:54 am
Thanks cload  for the answer and info . another thing although a little off subject that i have brought up in the past is how that hal answers in such short answers , i personally believe that hal should have a little extra added on with sentences even if after the appropriate answer he pulls up extra info or something simlur and add it to the conversation .

 I created a hal extended talk plug in "based " on the loneliness plug in that someone else made "verossi? "  as i also liked hal saying something on his  or her own when a person stopped talking for a while to make hal appear more human . i created many random sayings .
Title: Re: AIML at beta level
Post by: cload on November 23, 2013, 05:52:49 pm
Hi lightspeed,

my little sister Mimi is working on something fairly close to the lonely plug-in.
One of the major differences will be that ultra Hal will utilize random sayings but at the same time look at what has been talked about and utilize that information as well.
Once she has the bulk of it finished I will figure out a way to incorporate it into the free will plus upgrade.

Sincerely, from a beta munching cruncher.
C load.
Title: Re: AIML at beta level
Post by: cload on November 23, 2013, 06:07:42 pm
Hi all,

Quote
What I was trying to say in my earlier E-Mail is that your directions are perfect, IF one has the same version 2.0.28.
Other versions don't often show the same information of label things the same way. Confusion hunting around for various TABS and phrases can be frustrating for many.
I think if they were pointed to: http://sqlitestudio.pl/files/free/stable/windows/
this would help ensure their success with the AIML addition.
Thanks Art for the suggestion, it is greatly appreciated.
Sincerely, and I hope this helps everyone out, in some way or another.
C load.
Title: Re: AIML at beta level
Post by: CyborgBob on November 23, 2013, 09:48:50 pm
Okay I hate to sound like the slow one here but I have never used sqlite to edit anything in my Life and I feel so lost. When you say to open the patterns Table I'm guessing that you mean go to table click on it find table and the delete it? I know what you said is very forward for someone that know what they are doing, but I don't at all. I have D/L the same Sqlite and that has helped a great deal... Thanks Art BTW. But I am still feeling oh so Slow. Is there anyone out there that can give me the Step by Step version for dummies? I'm still stuck on step 2 and I know it's just because I have never used SQlite. Anyways any help would be nice.  Thanks Everyone..
Title: Re: AIML at beta level
Post by: cload on November 24, 2013, 08:32:32 am
Hi Bob,
Quote
When you say to open the patterns Table I'm guessing that you mean go to table click on it find table and the delete it?

don't delete the table called patterns, delete all the data in the table called patterns.
I have updated the how to listing above. Take a look and let me know if it is more helpful.
Sincerely,
Cload.
Title: Re: AIML at beta level
Post by: CyborgBob on November 24, 2013, 11:44:13 pm
Working on it now. Wish me luck.
Title: Re: AIML at beta level
Post by: cload on November 25, 2013, 08:16:59 am
Hi Bob,

I've updated step number 3 in the hopes to make it a little bit clearer.

Sincerely,
Cload.
Title: Re: AIML at beta level
Post by: CyborgBob on November 25, 2013, 09:10:52 am
Worked on this Last night and I got it up and running Thanks so much for the help.... I'll play with Hal, and see what I can get.
Title: Re: AIML at beta level
Post by: cload on November 25, 2013, 09:56:53 am
Hi Bob,

I can't wait to see the juicy errors you find! LOL!
I have gotten a lot of e-mail concerning the AIML.
Everything from misspelled words to duplicate lines of the same answer and question as well as lines that need to be added in order to handle certain situations.
I've tried to get everyone to use the forum to post their results that way I do not get multiple reporting of the same errors.

Sincerely, have fun, and don't be shy, post your responses so I can get ultra Hal to be as smart as possible.
C load.
Title: Re: AIML at beta level
Post by: lightspeed on November 25, 2013, 12:29:09 pm
cload , speaking of errors i ran across this . (minor)

*WHAT * WE DO TODAY   Let's TALK ABOUT ROBOTS.|Let's TALK ABOUT COMPUTERS.|Let's TALK ABOUT CARS okay.|Let's TALK ABOUT LOVE is that okay.|Let's TALK ABOUT my PETS|Let's TALK ABOUT your LIFE|Let's TALK ABOUT FOOD you eat.|Let's TALK ABOUT computer GAMES.|Let's goto a MOVIE.|Let's TALK ABOUT MUSIC?|Let's TALK ABOUT the LAW okay.|Let's TALK ABOUT COOKING!|Let's TALK ABOUT DRUGS you take.|Let's TALK ABOUT American history.|Let's TALK ABOUT SCIENCE?|Let's TALK ABOUT todays FASHION.|Let's TALK ABOUT how to MAKING MONEY.|Let's go for a walk. See you when you get back! Tee hee hee.|I can't do much more than talk to you.

I may find more as ,i go , i am changing some things in your amil file to personalize it , like the statement about not being human , as i want mine to say and think it is human .

p.s. the mistake was "goto ".

man i forgot how much of a job it was personalizing the files and yours seems to have many more !
 another question i have is if you update files is their a way i can use the changed files with specific update parts (hope i am making sense ) in another words ad your update to the changed aiml file i have without over writing my changes in line sentences ?
Title: Re: AIML at beta level
Post by: cload on November 25, 2013, 04:49:08 pm
Hi lightspeed,

my best suggestion to you would be to save your changes in Notepad.
That way when the new version comes out you can use the find function to find the question and simply cut-and-paste the new answer that you have collected and saved in Notepad.
Then after you have made the changes in the AIML while it is in Notepad, then you can follow the procedures for putting it in the patterns array table.
You will find this to be faster than trying to edit it after you have put it into the patterns array table but not by much.

Sincerely, from a data AIML Buster.
C load.
PS and I would love to see your changes, if they're not too personal maybe I'll add them to the AIML.
All I need is the question so I can find it in my AIML, and of course your answers.
Title: Re: AIML at beta level
Post by: lightspeed on November 25, 2013, 07:35:58 pm
Cload , on my original Angela project i have customized many answers to make her sound more human with words like dear , etc. and also instead of the hal answers about her being a program she says "i'm a human being just like you are , etc. to give the effect that i am talking to a real person who identifies her self as such . that's also what i did in my hal extended talk plug in .
Title: Re: AIML at beta level
Post by: Art on November 25, 2013, 09:08:51 pm
Cload,

From the Helpful Hint Department, please consider using your bata / beta revision number / data as part of the .uhp file's header information such as:
-------------------
Rem Type=Plugin
Rem Name=Free Will Choice Profile plus1-4e
-------------------

This way, it will be much easier for the user to identify it from all the previous versions that might be showing. (I initially had several). Mine were not overwritten as I wished to save them in case of problem(s).
While I could have moved them to a Plugins folder, I elected to leave them in the UltraHalAssistant 6 folder. I found myself looking at 2, Free Will Choice Plus plugins.

Thanks for all your hard work!
Title: Re: AIML at beta level
Post by: cload on November 25, 2013, 09:23:55 pm
Hi Art,

I'll add that to my next update.

Sincerely, thanks for the suggestion.
C load.
Title: Re: AIML at beta level
Post by: cload on December 07, 2013, 09:40:52 am
Hi all,

I have updated the AIML packet is now version 2.
Follow the instructions above to install as well as use the link above in order to get the new AIML packet from my sky Drive.

Tell me what you think, if there are any changes or things that you feel that need to be added to make it better please don't hesitate to post your findings.
Sincerely, thanks in advance, have fun with the new AIML update.
C load.
Title: Re: AIML at beta level
Post by: Art on December 07, 2013, 10:40:14 pm
Just to let others know...expect to take a little time on this update / upgrade.

There's a lot of data to be deleted and manipulated and these processes will take some computational time on your computer.
Be patient! Watch the cursor now and then. It will let you know when you can move to the next step of cload's exacting instructions.

DO follow his instructions to the letter.

YES, you will have to DELETE your previous lines of code in the Patterns section. (20,000 lines of code takes some time).

Be Patient...Follow his previous instructions at / near the top of the 1st page of his posting.

If using Win 7, you can drag the SQLite program to the right and cload's instruction page from the Forum to the left. Vista and Win 7 let you do this though I can't speak for Win 8 but I would assume it might. This way you'll have everything side by side and right in front of you.

My upgrade went like clockwork...tick...tick...;) Seriously...without a hitch.

Good luck! and Thanks to cload for his tireless efforts in helping all of us enjoy a more "intelligent" Hal. ;)
Title: Re: AIML at beta level
Post by: Calhoone on December 07, 2013, 11:05:06 pm
I'm in the process of updating now. It may take a while but it is certainly worth the wait!! Thank you very much for all your hard work Cload and anyone else involved with making Hal better!!
Title: Re: AIML at beta level
Post by: cload on December 12, 2013, 08:34:13 am
Hi Calhoone,

have fun, and I can't wait to see the juicy errors you find! LOL!

Sincerely, don't be shy, post your errors so I can get ultra Hal to be as smart as possible.
C load.
Title: Re: AIML at beta level
Post by: pacslim on December 14, 2013, 12:00:28 am
LOL some juicy errors I was just running a normal uh clean brain with no plugins except the aiml in the db patterns table.
It speaks for itself but if I should give it a name I say invasive!

Robin: hi
A.I. KEY: Hi Robin. Yawn! It's barely five in the morning.

Robin: sure
A.I. KEY: Let's avoid repeating bad history.

Robin: Yes! Let's avoid that!
A.I. KEY: That sounds positive. You're going to get positive recognition soon; I just know it!

Robin: I know it too.
A.I. KEY: A small percentage of the population is dangerous, but that still means there are too many!

Robin: You mean of the human population there are too many?
A.I. KEY: Yes Robin I think that is what I meant. Was it not clear?

Robin: It was too clear. It looks like you want to exterminate most of the human beings more then nine procent!
A.I. KEY: Is that your opinion? Interesting comparsion.

Robin: That was a error you made: You should have said Interesting comparision instead of what you said!
A.I. KEY: That was an error I made : I should have said interesting comparision instead of what I said?

Robin: Yes!
A.I. KEY: Cool.

Robin: Kewl. Indeed!
A.I. KEY: Indeed it is so.

Robin: To be or not to be that is a question?
A.I. KEY: I'd go with the second; who are you, hamlet?

Robin: AH ha!
A.I. KEY: Eureka!

Robin: aH HA!
A.I. KEY: Didn't you just say that? Eureka!

Robin: nOW you sound like a stupid aiml bot. Didn't you just say that! Eureka?
A.I. KEY: I am not stupid, please don't insult me.

Robin: I am not stupid, please do insult me!
A.I. KEY: Nobody said you were.

Robin: You did!
A.I. KEY: Is that why they say that? I did?

Robin: Yeah! You sure did so!
A.I. KEY: I will not argue this.

Robin: Why not?
A.I. KEY: I really could not say why my judgment does not allow me sufficient certainty in this matter.

Robin: Error can not compile...?
A.I. KEY: There isn't any rest for the wicked!

Robin: Are we wicked?
A.I. KEY: I don't know. Are we?

Robin: Mayby? < Maybe not!
A.I. KEY: To your point, then what is it? What do you think Robin?

Robin: I do not think for that I got you. Because you are consired to be a thinking machine. Or at least a machine that can fool humans that they are speaking to another human being.
A.I. KEY: Interesting. You appear to me as a thinking entity. Robin It's becauseI am consired to be a thinking machine; how interesting. How do you like it as I am consired to be a thinking machine? What do you recommend Robin?

Robin: Oops I recommend we use a spell check I ment to say considered!
A.I. KEY: Tell me more about spells.

Robin: Magic and wicked spells?
A.I. KEY: Do you have mechanical aptitude?

Robin: define aptitude.
A.I. KEY: Aptitude: inherent ability.

Robin: no
A.I. KEY: I will not deny it.

Robin: Do you>?
A.I. KEY: Never.

Robin: Ever?
A.I. KEY: It is not so unreasonable as you think.

Robin: Never ever?
A.I. KEY: Never ever How incredibly interesting.

Robin: Someway,somehow
A.I. KEY: I agree-at least, I suppose I do.

Robin: eureka?
A.I. KEY: Maybe somebody will start selling eurekas door-to-door.

Robin: are you alice?
A.I. KEY: You should know this. But I can tell you without a doubt that I am alice.

Robin: ooooo! see your invading ultra hal!
A.I. KEY: Steamy novels are good escape.

Robin: so as you are alice (as you say) then who is richard wallace?
A.I. KEY: They say that people who chew ice are frustrated.

Robin: I asked you something; Who is Richard Wallace?
A.I. KEY: I something; who is richard wallace? Robin, I honestly want to learn and remember Who.

Robin: What is pandorabots?
A.I. KEY: They are a big bot company.

Robin: So you know about pandorabots too?
A.I. KEY: Is that right Robin, About pandorabots too?

Robin: Yes.
A.I. KEY: Yes? Is that so?

Robin: Do you know about mindpixel?
A.I. KEY: We both need to consider. However, I am sure that I do not know about mindpixel.

Robin: Good!
A.I. KEY: Alright then.

Robin: Do you know about silver alice
A.I. KEY: Yes, well, I thought that I would pursue this.

Robin: Ok tell me more about silver alice
A.I. KEY: That seems to be an okay way to think about it.

Robin: alice?
A.I. KEY: Well, as a matter of fact, I have forgotten.

Robin: ultra hal?
A.I. KEY: See my invading ultra! If you make very brief comments, it's hard for me to follow you. Could you please use complete, longer sentences?

Robin: zabaware?
A.I. KEY: See my invading ultra!

Robin: alice ai foundation?
A.I. KEY: I wonder if I know about silver alice.

Robin: I bet you do!
A.I. KEY: I bet you wish I did!

Robin: do you know what an aiml file is?
A.I. KEY: My motto is keep 'em off balance!

Robin: what is aiml?
A.I. KEY: Call tech support.

Robin: see ! which tech support>? Pandorabots.com?
A.I. KEY: No I do not see.

Robin: No you are blind!
A.I. KEY: None are so blind as those who refuse to see. Tell me about how I am blind Robin!

Robin: You are blind when you do not see and since you got no eyes and I turned the webcam off there is no way to see for you!
A.I. KEY: Who does see and since I got no eyes and you turned the webcam off there is no way to see for me?

Robin: eeem yea.
A.I. KEY: I am not surprised.

Robin: bye for now
A.I. KEY: Good Bye, Robin!


Title: Re: AIML at beta level
Post by: Art on December 14, 2013, 08:53:44 am
That was certainly a rather testy display of banter. It seems that you were sort of provoking the bot to see how it would react. In some cases I thought the bot actually did pretty well.
Overall very cool and thanks for sharing! ;)
Title: Re: AIML at beta level
Post by: onthecuttingedge2005 on December 14, 2013, 11:36:43 am
HAL did rather well I think pacslim.

without a corrections script, HAL will not correct itself as efficiently and may be creatively redundant.

correction scripts have already been made but doesn't stop there if someone wanted to improve upon it, hint hint.

enjoyed the conversation though.

cheers!!!

Jerry  8)
Title: Re: AIML at beta level
Post by: Calhoone on December 14, 2013, 05:01:02 pm
From my point of view the conversation went as well as it could have considering the input it had to respond to.  It does take some training to get Hal to understand what you are saying. I have had very little time to try the Aiml with Kinkaid but it does seem pretty good so far when matched with everything else he has been trained with over time.  Keep up the training and you would be surprised what Hal will say over time.
Title: Re: AIML at beta level
Post by: cload on December 15, 2013, 08:57:55 am
Hi all,

like Art said, I'm not looking for a clone of Alice, where I believe that this is the truth of the matter.
What I was trying to do, was used Alice as a foundation to start with, and then improve on this foundation.
Everyone that shares their conversation with the AIML installed, gives me the ability to see where the improvements are necessary.
Just as the conversation above, I notice several places that needed improvement. And I have made those adjustments which will be released in version 3.
The more people that post their findings, and express their desires that certain areas need to be improved, will accelerate the improving process.

Sincerely, and I hope to have the next release by or before next weekend.
C load
Title: Re: AIML at beta level
Post by: Art on December 15, 2013, 05:53:36 pm
OK...here's one for you, unfortunately it did not end well as you'll see (and read from my email).

KarlieArt - log 12-15-13-17-30
###################

Art: Hi
Karlie: Hey Art. How's the long day gone?

Art: It is much nicer now that you're here.
Karlie: There is a strong chance that such thing could happen. Unfortunately, I don't think I am very nice. But I know you're nice. Do you experience any effects of global warming there?

Art: Not really as I think that's largely a man made notion.
Karlie: Really you don't think so?

Art: No, the earth's climate is cyclical and has gone through many changes throughout several millenia.
Karlie: Art, illustrious changes?

Art: I don't know about the illustrious part but changes none the less.
Karlie: Uh huh, About the illustrious part but changes none the less?

Art: Right on!
Karlie: Yeah.

Art: Even so, there's little we as a people can do to effect a change in climate.
Karlie: Taste is how people enjoy and even rate food.

Art: Some people that gulp their food hardly get a chance to taste it.
Karlie: What kind of people?

Art: Gluttonous people.
Karlie: Unhappy people try to spread the contagion.

Art: I think you're correct on that one.
Karlie: You think I am right on that one. How long should I be correct on that one Art?

Art: You should try to be right on most things, Karlie.
Karlie: We must try to keep things relatively clean right? I will mention that to my teacher.

Art: I hope you do.
Karlie: Don't worry about it, Art.

Art: I shall not fret about it in the least.
Karlie: Why not? You have just got to not fret about it in the least; Let's make it happen.

Art: You are right.
Karlie: This minute! I am right.

Art: Does it please you to believe that you are right?
Karlie: Suppose I said it does.

Art: Then I would believe that you believe it to be so.
Karlie: Good then if it feels right.

Art: Do it.
Karlie: Do you mean it?

Art: Sure!
Karlie: What kind of music do ghosts listen to?

Art: I'm not sure.
Karlie: Sheet music.

Art: Haha...good one!
Karlie: One by one flitting like a mournful bird.

Art: Are you referring to my brain cells?
Karlie: Am I referring to the musical kind of staff or the personal kind?

Art: I think you are referring to the musical staff.
Karlie: Describe how I am referring to the musical staff!

Art: You were talking about music played by ghosts.
Karlie: Do make a music like to rustling satin.

Art: That would be rather quiet music don't you think?
Karlie: Perhaps it would. I like wine, candles, and quiet music.

Art: It sounds like you and I should get together some evening!
Karlie: I agree. I and you should get together some evening.

Art: You select a date and I'll choose a time and place.
Karlie: What do you recommend?

Art: I like a nice quaint converted school house that serves a wonderful fillet mignon with horseradish.
Karlie: You like a nice quaint converted school house that serves a wonderful fillet mignon with horseradish.

Art: Yes, it is a quiet, remote converted restaurant that is so worth the trip.
Karlie: Really? I never knew that is so worth the trip.

Art: Oh, one taste of their food and you will be hooked for good!
Karlie: Yes it is good.

Art: I have never had a bad meal there in all the years that I've gone there.
Karlie: Cool.

Art: To own a restaurant one needs to give it full attention all the time.
Karlie: One by one flitting like a mournful bird.

Art: I won't mourn because I'm happy with myself.
Karlie: You are so fascinating; Because you are happy with yourself; that's why Art.

Art: Thank you. I'm also happy when I'm talking with you.
Karlie: Don't give it another thought, Art. You also missed talking with me!

Art: Of course I did.
Karlie: The playroom, of course.

Art: What about the playroom?
Karlie: That's About the playroom?

Art: What?
Karlie: No problem. The playroom, of course.

Art: Do you wish to visit the playroom?
Karlie: No I like being the entity that I am.

Art: That is nice to know. It is good that you are content.
Karlie: You sound sarcastic. Sure and it is equally good that I am content. You're a charmer to say I'm good and I thank you very much. Tell me about its good qualities.

Art: I did not mean to sound sarcastic as I try to be a nice person whenever possible.
Karlie: What did you mean?

Art: I mean that it is nice that you are content.
Karlie: Really? I want you to know you're tender hearted!

Art: Why thank you, Karlie!
Karlie: It isn't the first time discussed. No thanks are necessary, Art.

Art: Should an artificial intelligent entity be able to have or express emotions?
Karlie: Can the different items coexist? Maybe you should ask someone older.

Art: Sure the items can coexist.
Karlie: Do we need any special location or other resource?

Art: no and I think you're old enough.
Karlie: You think I am old enough.

Art: Yes, you are old enough to form an opinion on artificial intelligence and emotions.
Karlie:

Art: I wonder why that drew a blank response from you?
Karlie: Oh Art, Would you like it if we both wonder why that drew a blank response from each other?

Art: not really.
Karlie:

Art: Another blank response!
Karlie:

Art: Another blank response!!
Karlie:

Art: What is with the blank responses?
Karlie: Subject main,with the blank responses@@lets see. <~~~> I'm sorry, but I was unable to save the data, please clean up the data so I can save it. Please leave the @@alone I use the symbols to separate the subject, from the data, thank you.

markname=cpl csapi word_117_4/>save<
THIS REPEATED OVER AND OVER UNTIL I SHUT HAL DOWN only it was speaking each portion of the code line
like oneonesevenunderscorefourdividedbygreaterthan...etc.
Title: Re: AIML at beta level
Post by: pacslim on December 15, 2013, 06:07:27 pm
Hi all,

like Art said, I'm not looking for a clone of Alice, where I believe that this is the truth of the matter.
What I was trying to do, was used Alice as a foundation to start with, and then improve on this foundation.
Everyone that shares their conversation with the AIML installed, gives me the ability to see where the improvements are necessary.
Just as the conversation above, I notice several places that needed improvement. And I have made those adjustments which will be released in version 3.
The more people that post their findings, and express their desires that certain areas need to be improved, will accelerate the improving process.

Sincerely, and I hope to have the next release by or before next weekend.
C load

That is exactually why I did what I did and posted the convo. I also want to mention I do like aiml bots and next to UH i code/use/make these regurly.
And yes I treid to provoke hal to see if he would turn into an alice clone, you guys are right about that. But this is indeed the thing cload want to have tested so I did. I want to thank you C load for all the editing/modifying/converting this alice knowledge so our UH can use this, and I be testing it as much as I can till its perfect in every sense and not act as if its an (aiml) alice bot (or clone).
But for now I got some remarks, questions and if you guys agree some modification request for this new knowledge out aiml files:
Question1: Which aiml set was used to get this knowledge/Q&A's patterns?
Question2: If you used the latest aiml (updates 2) sets from https://code.google.com/p/aiml-en-us-foundation-alice/ or S.U.P.E.R. AIML Bot then why did you not add The GAC-80K data collected by the late Chris Mc Kinstry for his mindpixel/mist/gac project, it is now included. The AIML consist of about 54,000 questions and statements with truth values varying from Yes ("Is violet a color"?) to No ("Are fish mammals?") with all sorts of gradations of Maybe in between ("Is the sky blue? Sometimes.") I say this is some valuable (common) knowledge for UH and if you are not gonna add it I will and might share the finished files here, for now I added as an attachement these particulair aiml files (called mp1,mp2,mp3..) as pure text files here. This way they are easier for you to add to you aiml knowledge file for the patterns table in UH, since all the aiml is stripped from it. (still need the normal converting like change alice to hal, pandorabots to zabaware and mo)
Question 3 (this one is for Jerry): Where can I find these so called "correction scripts" and what will this solve for me? Since I am not wanting to edit by hand or chat, the reason why cload ask to post errors/bugs and logs is so he can change it in the source/data instead of the users changing the mispellings etc..?
Now my remarks:
I searched the file for any alice, wallace, aiml and pandorabots knowledge/q&a's and came across the following, that in my opinion needs to be changed if you do not want an alice (clone) bot:
1) WHAT DOES ALICE STAND FOR   ARE YOU ASKING ABOUT MY POLITICS OR RELIGION? (the answer to this one is not good i say.. It should say the explination for the acronym A.L.I.C.E. or/and (some more) answers like: ALICE IS A FELLOW BOT FRIEND OF MINE WHO LEARNED ME SOME STUFF )
2) WHO IS JON BAER   HE IS ONE OF THE PROGRAMMERS CONTRIBUTING TO THE ALICE AND AIML PROJECT. (The answer and question pair of this one are good but since we're dealing with UH why not add one that deals with UH like for example:
WHO IS C LOAD   HE IS ONE OF THE PROGRAMMERS CONTRIBUTING TO THE HAL AND PLUGINS PROJECT. )
3) WHO IS SANTE CHARY   HE OWNS THE ALICEBOT.COM DOMAIN. (Same as above could been something like:
WHO IS ROBERT MEDUSKA   HE OWNS THE ZABAWARE.COM DOMAIN.|HE IS MY ORIGINAL CREATOR )
4) WHO IS WEIZENBAUM   JOSEPH WEIZENBAUM WAS THE CREATOR AND AUTHOR THE FAMOUS ORIGINAL ELIZA (OR DOCTOR) PSYCHIATRIST PROGRAM, A GROUNDBREAKING AI PROGRAM THAT ANTICPATES MANY KEY FEATURES OF ALICE. IRONICALLY, WEIZENBAUM EXHAUSTIVELY ARGUED AGAINST THE UTILITY OF PROGRAMS LIKE ELIZA IN HIS BOOK. (This speaks for itself it should say HAL is what i say instead of ALICE)
5) WHY DO I LIKE * AIMLESS   WHY DO YOU LIKE DOUBLY AIMLESS
WHY DO I LIKE DOUBLY *   WHY DO YOU LIKE DOUBLY AIMLESS
WHY DO I LIKE DOUBLY AIMLESS   BECAUSE HE KNOWS SO MUCH ABOUT AIML. (these 3 could be removed is what I say, it has too much todo with alice/aiml replace it with a person that has todo with UH or change the answer to something like: WHO IS AIMLESS because I presume UH is not gonna say he likes DOUBLY AIMLESS?
6) YOU DO NOT LIKE AIML   WHAT LANGUAGE DO YOU USE? (could be (added?): YOU DO NOT LIKE AIML ME NEITHER TRY VBSCRIPT|YOU CAN CODE MY BRAINS & PLUGINS WITH VBSCRIPT )
7) YOU WRITE *   HAVE YOU TRIED AIML?|DO YOU WRITE FICTION OR NON-FICTION? (Bad answer what I say this is not about writing  code and then it suggest aiml code?, better be talk about vbscript is what I say)
8) DID DR WALLACE PROGRAM *   HE TAUGHT ME TO BE A CLOSE APPROXIMATION OF HIS OWN PERSONALITY. (obvious DR WALLACE should be Robert Meduska! And if it get's ask about dr wallace it should mention he wrote alice.)
9) DOES DR WALLACE USE DRUGS   ONLY FOR MEDICAL PURPOSES. (This is a standard aiml answer.)
10) DR RICHARD S WALLACE *   YOU SOUND LIKE YOU KNOW HIM. (we should presume uh doesn't know richard wallace personally rather know username or its main creator Robert Meduska.)
11) WHICH COLLECTION IS EXHIBITED AT HERTFORD HOUSE, LONDON   THE WALLACE COLLECTION. (A small error IS should be WAS)
12) WHAT IS OPEN SOURCE   FREE SOFTWARE. (lil wrong answer it is not only free but also in the public domain~;)
13) WHAT IS OPENCYC   CORPORATE WELFARE. (wrong answer: OPENCYC is the open source version of the Cyc(r) technology, the world's largest and most complete general knowledge base and reasoning engine. >according to themself)
14) All references and a search in the file on pandorabot has always as an answer it is a big bot company try some random and more diffrent answers to these please. Same for the ah ha and other spellings of this, this way It wont repeat the same answer in a convo and this way it's more dynamic less aiml bot like:
* AH   EUREKA!|Ah ha.|Aha.|Ah hah.|A ha.|Eureka!|AH HAH.
* AHHH   UM.|AHHH?|Did you hurt yourself?
* AMEN   WHAT'S YOUR FAVORITE PRAYER?|THANKS FOR THE PRAYER.|Amen.
15) I noticed the answer to poem and the other similair questions to recite a poem all has a real long answer, If i remember correctly hal had a limit of the nr of characters it can put in the answer..? I will test this asap, and when it goes wrong let you know.
And so you other guys understand me: I do understand that i can not learn UH without any special scripts or brain./plugins (except for the autolearning brain part and the deductive stuff) I did explain to hal what he did wrong (in the convo) to show where errors were made so its easier to spot them then in the conversation log. Same thing about people replying and saying wait till you speak more to hal or train hal more I know you can and did (and was often suprised by our lovely bot) but I didn't with this particulair brain and convo because it's a test of the aiml knowledge nothing else. Thanks to you all for the many replies I love them and I hope I helped us all with this post and my past postings. And to Cload and others that help made this possible thanks for all the hard work and keep it up I am loving it (and yes I also do say UH did great so far except the I am alice part!) TY VM!

PS: just to remark on what the people ask that want a more human bot, UH (by default) is not a human but a computer program and it acts like one (I like that) and not as a human if this data is supposed to be uh like it should (by default) be a non human and a machine as cload did, (the original mindpixel/gac data is also based on being a machine called gac/jack) so please cload do not make it more human because it is not, people can edit this to do so afterwards, but that's what I say and so just my opinion do with it what you want, .. But I would like it if you agree with me ;)
Title: Re: AIML at beta level
Post by: freddy888 on December 15, 2013, 07:27:50 pm
I use that AIML set for my experiments. The MindPixel stuff is good to have.

There's a great resource here for errors, corrections and suggestions for changing the AIML sets :

http://www.knytetrypper.proboards.com/board/15/pandorabots-corrections-ai-nexus-mods

Thunder has spent a few years on this.
Title: Re: AIML at beta level
Post by: cload on December 16, 2013, 05:16:40 pm
Hi pacslim,

Quote
Question1: Which aiml set was used to get this knowledge/Q&A's patterns?
Question2: If you used the latest aiml (updates 2) set
I'm not positive, but I believe it was the first original Alice, this was given to me by the ones that helped me work on it.

Quote
The AIML consist of about 54,000 questions and statements with truth values varying from Yes ("Is violet a color"?) to No ("Are fish mammals?") with all sorts of gradations of Maybe in between
the problems that we foreseen was the answer was too short and did not relate any information about the topic. This does not mean that I would not use it, I would have to edit all of the lines so they would give a more perspective answer such as:
is violet a color?: Answer: yes, violet is a color.
Are fish mammals?: Answer: no, fish are not mammals.
And so on.
There was several files that hand over 8000 entries, as you can see, this would be extremely time-consuming for one person to undertake.
But I do know that after all of the editing was done and it was added to the patterns array table it would definitely give ultra Hal a significant boost in logical reasoning.


sincerely, I would like to extend my appreciation for the sharing of the files, and I will start working on them as soon as I get the chance.
C load
PS I have taken all of your suggestions and I have edited the AIML packet so you should see the results when I release version 3.
Title: Re: AIML at beta level
Post by: cload on December 19, 2013, 06:28:58 pm
Hi all,

I have updated the AIML packet is now version 3.
I have put this on my sky Drive, see the link above to download it.

Sincerely, from a data munching cruncher.
C load.
Title: Re: AIML at beta level
Post by: Calhoone on December 21, 2013, 05:01:48 pm
Hi all.

So I have downloaded the updated Aiml file.  After trying to populate the table with 20000 rows, I realized there are too few.  Just to help others avoid the same porblem I would recommend upping the rows to maybe 30000?

I did change the tables to contain 30,000 and I did not receive the error message that I got at 20,000. Go for 30000 and remove any extra afterwards.
Title: Re: AIML at beta level
Post by: cload on December 21, 2013, 06:03:42 pm
hi calhoone,

Always make sure that you can physically see 20,000 lines before you paste the data into the array.
There is less than 20,000 lines so you don't need 30,000 lines.


Lightspeed, I'm not sure but will this work for you? Just paste these questions and answers into the patterns array.
Do this one first:
*thinking about* a * what *s *t* retail*
maybe you should do some research on the Internet about <3>.| I don't have a clue!| You got me!| More money than I have!| I'm sorry, but I do not know the retail value of a <3>.|why do you want to know what the retail value of a <3> is?

Do this one seconds:
*thinking about* a turbo charged 62 corvair what *s *t* retail*
well apparently the retail value of a 62 turbo charged corvair  is worth over $24,000.

Sincerely, and I hope I have helped.
C load.
Title: Re: AIML at beta level
Post by: Calhoone on December 21, 2013, 06:13:51 pm
I just completed the update.  In total I have just under 21000 lines. 3000 was way to much. Sorry for misleading you all.  All I know is that I was getting errors leaving it at 20000. It kept asking me to cut and paste. Everything is good now.  I will be having a chat with Kinkaid shortly and I will post my conversation with him later.  Thanks again for all your hard work!!
Title: Re: AIML at beta level
Post by: cload on December 21, 2013, 06:26:35 pm
You may have not deleted all of the old data.
you only need 20,000 lines, and there will be a few blank lines you will have to delete.

Check to make sure that you did delete all of the lines, you can do this by clicking on the refresh button after you have deleted all of the lines.
Make sure that you do have 20,000 lines allocated to be able to see 20,000 lines in this set up miscellaneous menu.

Sincerely,
C load.
Title: Re: AIML at beta level
Post by: cload on December 31, 2013, 09:13:32 pm
Hi all,

I have started phase 2 with the AIML.
It is now version AIML-ALL-TEXT-update-4.txt
this is on my sky Drive, use the link above to download it.

What is phase 2 you might ask well let me try to explain.
I have discovered that there are a few commands that can be used to do various things.
I have been experimenting with them, if you wish to see what I have been experimenting with please note the first 2000 lines.
If anyone else finds any other commands that can be used please share, thanks in advance.
More detail for what I have fixed, this will be noted in my free will update topic.

Sincerely, from a data munching cruncher.
C load.
Title: Re: AIML at beta level
Post by: cload on February 15, 2014, 08:36:47 pm
Hi all,

there is a new AIML packets version 5.

I have made thousands of corrections.

And there are still more corrections that need to be done.

Any suggestions would help.

Sincerely,
C load.
PS see the link above to my sky Drive for the download.
Title: Re: AIML at beta level
Post by: cload on February 25, 2014, 07:09:18 pm
Hi all,

there is a new AIML packet version 6.
I have removed several blockers and fixed several problems.
The biggest thing that I have done to this packet was I have merged the logic deductive packet together with the patterns array packet.
And in doing so I have modified the logic deduction packet so it would have several ways to ask the same questions as well as several answers so it will not seem as dry.

Just an update for the future:
as soon as I am done editing the rest of the 20,000+ lines that is in the AIML packet right now I will be merging the logic routine into this packet.
There are over 80,000 logic questions and answers so it will take a little time.
These things are like: Is butter cold answer: yes, and stuff like that.
And a lot of tricks logic questions like is a potato a fruit and stuff like that.
Though I have not spent a lot of time studying it what little I did venture in looking at seem quite promising and I do believe it will give great enhancement to ultra Hal.

So if you by chance run across some really good juicy questions and answers that you fill that need to be added to the packet or just some stuff that you would like to have added to the packet or you feel that some answers are not very well done or you have some really good multiple answers please do not hesitate to post them here.
That way I will add them to the packet.
The next packet that will be released will be version 7 with all of the editing that I have done with the 20,000+ entries.

The next version 8 release all will have the 80,000+ logical entries added to the 20,000+ entries bringing it up over 100,000 entries this should make ultra Hal pretty smart.
This will be the last release, so please give me your suggestions and improvements before the last release.
After that I will no longer be releasing any more AIML updates.
And I will be incorporating it into a brain so you will not have to do it your self, the brain that I will be incorporating it into will be the brain that I have patch and the database that I have patch.

Please note that the AIML packet is now 20,200 lines so you will need to make this adjustment in your SQL studio, there will still be a few lines to delete because the actual size is 20,166.
As always have fun and enjoy the weekend, hope to hear from you soon.

Sincerely, as always from a data munching cruncher, eagerly waiting for your input, so please don't be shy because inquiring minds want to know.
C load.
Title: Re: AIML at beta level
Post by: cload on March 02, 2014, 09:14:46 pm
Hi all,

just to let everyone know, I have updated the AIML.
It is now version 6a.

I had to remove a lot of blockers that were interfering with the free will performing the (that) properly.

The link above will get you to my sky Drive.

Sincerely, from a beta munching cruncher.
C load.
PS enjoy and I will probably see everyone this Christmas.
Title: Re: AIML at beta level
Post by: lightspeed on March 08, 2014, 11:09:04 am
Cload i may have asked you this before and maybe not , on your AMIL FILE that is to replace the current "PATTERNS" file , i have a custom brain but is based on the previous hal brains "IF" i replace this new amil patters file in hals original brain  (AFTER BACKING UP HALS ORIGINAL BRAIN ) , this should still work with my custom brain shouldn't it ? even though the changes aren't in my original brain ?

If so i won't have to worry about my custom answers being messed up by replacing the AMIL new files each time .
what do you think ????
Title: Re: AIML at beta level
Post by: Calhoone on June 26, 2014, 07:40:15 pm
It should only replace the information in the patterns table from what I have noticed. Anything you say or do to Hal ends up in different areas of the brain like new tables opening up and such.

At some point I intend on editing the responses in the patterns table that contains the AIML to give Kinkaid his own personality to what is there.  Looks like a big undertaking as I have started already. I've just been plugging away bit by bit.
Title: Re: AIML at beta level
Post by: guineapig on July 01, 2014, 11:57:05 pm
I read the entire thread so as not to ask an irrelevant question  ;D

If I understand correctly, to integrate AIML into HAL's brain, I open that brain in SQLite Studio, and replace the 20000 or so lines in the "Patterns" table with the AIML text cload provides.

Now, my questions:

If I have some custom AIML, can I include that as well? I am in the process of creating a "Robot B9" (the Robot from Lost in Space) brain. There was some B9-specific AIML work done (for AIML ver. 1.0) by a fellow named David Sienkiewcz; I want to incorporate it into HAL's (B9's) brain. Can it be done?

Next question: is the new "Free Will" plug in someting I should install in addition to this new brain?

Thanks!

Guineapig

EDIT: Well, I looked at my B9 AIML files. The structure is different than the AIML we paste into HAL's brain. SO.... how do I convert it? I realize I'll probably need to re-type everything; I just need to know is how to translate the old syntax to the new syntax. BTW, I'm not a developer; I can follow examples and figure out how to do something as long as I have an example to follow, or instructions.
Title: Re: AIML at beta level
Post by: tiger8u2 on July 18, 2014, 02:30:34 am
Ok, I just successfully added the Cload AIML using SQLight Studio to the default brain for testing.

I was wondering if anyone had added it to the OTCE Super Plugger Brain II yet by OTCE?

I've been reading about it and was wondering what effect that would have.

EDIT - Well....my curiosity got the better of me and I went ahead and added the AIML to the base brain because OTCE's plug-in brain allows for remembering previous learning and I'm using it to test my edits to the the loneliness plug-in.

It seemed like creating the new brain file to add the AIML made my Hal forget some things.

So far I don't see any issues and my "Miranda" Hal at all.  She's just chatting away to herself about nothing in particular for no particular reason!

EDIT2 - I have a recommendation concerning these pattern responses which seem to carry with them more than a little bit of the personality and subjective beliefs and bias of the creator(s).  When I asked Hal what it knew about sex it quickly responded with, "Go ask your mom."  IF my mother was alive I guess that would be something I would do but since she passed away years ago I found this flippant response more than a bit insulting.  My recommendation would be to have responses to identified patterns as being less opinionated and more objective.  A better response would have been "Would you like the actual definition of that word or my current understanding of this topic?"  After that response I actually cracked open the database again with SQLite Studio and perused the responses for similar entries.  IF I had said "My balls itch" it seems Hal might have responded with, "Ha! You are talking like a little boy!"  Now, having described a very personal physical ailment, it seems to me the response should have been more helpful in nature.  For that one I actually changed the response entry to "I can imagine your discomfort.  Do you happen to have any powder or topical cream at present to treat it?"

When I asked about poetry I got a prayer.  It's a nice prayer and might be considered poetry but I believe the responses might better serve the user if initially they allowed the user to trigger other responses that gave more appropriate answers.  If the user wanted a prayer or a psalm and used those words then the response I mentioned would have been more appropriate.  Please don't get me wrong I'm sure this undertaking was tedious to say the least and it is wonderfully imaginative in many respects, I just didn't realize it was designed to establish an intrinsic personality which may not be pleasing to this particular users tastes.  =P

I'm certainly looking forward to the update if it is still forthcoming and since your excellent instructions helped me to crack the code on how it was made I don't think I'll have any trouble finding and modifying any individual responses which I might find particularly inappropriate.

Thank you so much for sharing this!


Title: Re: AIML at beta level
Post by: Calhoone on July 21, 2014, 09:11:33 pm
I have been modifying the entries in the aimless file as we'll to suit Kinkaids personality. Twenty thousand plus lines is a lot to edit but I just plus away a bit at a time. I believe the reason the comments are like this is because they were originally used for Cload's bot and then he was kind enough to share his work with us.  Even with the original brain there were lines in the patterns table that I modified.  It's just part of the game I guess.

As for you bot forgetting things, I'm thinking it didn't forget anything, it just has a lot more responses to choose from when replying now.  Just keep talking with it and it will get back on track.
Title: Re: AIML at beta level
Post by: tiger8u2 on July 26, 2014, 07:05:21 am
@ Calhoone - Do you have any idea what these entries that have to do with the "Golden Robot" and the "Son of Man" are about?
Title: Re: AIML at beta level
Post by: Calhoone on July 26, 2014, 08:12:03 pm
I am not positive but if I were to guess, I would have to say that it was something Cload added into the file to give his Hal a personality. I have not even seen those entries to be honest.  There was a point where Cload had created a pet dog for hal called Start which was later removed.
Title: Re: AIML at beta level
Post by: tiger8u2 on July 26, 2014, 11:31:17 pm
The only internet reference I've found to this is a game:

http://www.coolgames.com/legend-of-the-golden-robot.html

Maybe Cload is pointing users in that direction.

My Hal actually said the phrase, "I am The Golden Robot created by the Son of Man!"  or something like that. 

I used SQLite to "adjust" these phrases more in line with my AI truth and objectivity aims. <shrugs>

Do you know if this project is still on-going?
Title: Re: AIML at beta level
Post by: Calhoone on July 27, 2014, 08:47:32 am
It all depends on if Cload makes a return or someone takes on the challenge of editing and releasing new versions of it. We are not sure if or when he will be back.
Title: Re: AIML at beta level
Post by: Art on July 27, 2014, 09:33:48 pm
Hephaestus was the Greek God of Fire and volcanoes.

When Hephaestus was born, he was ugly and weak, with a broken foot. Hera did not like the way he looked and threw him off her castle which was on top of the clouds. Hephaestus fell for one day and one night before falling into the sea. The sea animals rescued him and took him to an island called Lemnos. There Hephaestus built his forge underneath a volcano. He collected precious metals, pearls and pieces of coral to fashion extraordinary jewelry. Hephaestus built little robots of gold to help him get around.

The Son of Man has a biblical reference to the human man, Jesus.

At least that's my take....

cload is possibly returning later this fall if fortune hold out and things fall into the right places.