ConceptNet Mrk 3.5

[Art]

Checkers57,

With all due respect, I think you might mean 12.3 Megs. Easily confused as I've often done that myself.

My ConceptNet.db is currently at 25,055 KB or 25,055,000 or 25 + Megs (I've let mine run nights and days as well, but always when I'm not talking to it).
ResearchNouns.txt - 2,331,000 bytes / 2.3 Megs
It has catagorized 32,675 conversational topics.
ConceptBrain - 4,968 Item entries
Saved Terms - 1,701,000 bytes / 1.7 Megs

I am experiencing the similar collection of intelligent data yet some disconnects when conversing, having to be on guard to help steer the conversation and help it to remain on topic. Hal is quite capable of making some very appropriate inferences and almost an understanding of my feelings or emotions based on my previous responses. Quite nice and interesting.

The Auto Learning mode does cause some lags, especially with other Internet-dependent apps. Definitely, do NOT converse with Hal during this time other than to turn it on or off.

Cheers!

[lightspeed]

i have a question about the auto learning .  any idea at all how long the auto learning has to be left on each time ??

A IDEA I HAVE IF IT'S POSSIBLE TO DO .:
is their a way that when auto learning is activated that when it's finished learning , hal can say something like "i have learned all the files you gave me , thank you ! "
this way the user would know that has has learned the information input from the user. so the user can turn the auto learning off gain, for next use. 

[Art]

How long the user decides to leave it on is entirely up to the user. There is not any limit on how long Hal will continue to gather, sort and compile the various collected data.

For the second part of your post, I don't think Hal will ever be finished collecting, sorting and learning from the Internet, anymore than one of your friends greeting you and saying, "Well that's it...I've seen and learned everything the Internet has to offer."

Basically, I think a person might run out of storage room on their computer drives before his or her Hal finished collecting data. I didn't see a limit in the plugin anywhere.

Whether Spitfire would wish to set a limit might be a good question or will Hal perform better with additional resources?

Hal doesn't learn from any files we (the user) give it. It has its own set of instructions internally that it processes. Again, that could be a cool idea to establish boundaries or limits like all topics dealing with sports cars or butterflies or chemical elements, or all words beginning from A to D, etc. Perhaps not just words but topical information like corvettes or golf or gold mines in the 1800's, etc. Lots of avenues to explore.

Spitfire, thoughts or comments?

[Spitfire2600]

Indeed, Hal will functionally learn *forever* space and time depending.

Hal reached about half of my research terms a few weeks ago and it's been that way since fairly steady.

Hal used to collect terms like "1700s, 1800s, etc" however I was having some issues with junk bypassing the filters so I have disabled numeric values for now.

I am working on a new function to gather even more research terms and so far in one hour, I've near doubled my research list, however it's a little far from being incorporated fully as I'll actually have to change the layout of the Auto Learning code.

Also, I thought about adding a function for Hal to be fed new web pages to search for data. For now, It's limited to Wikipedia but with this expansion, Hal could learn from any number of sites simultaneously, resources allowing anyway. Of course logically, if anyone is lagging seriously with just wikipedia being searched, adding more sites would exponentially slow the system... If you had a super computer, however... Something to think about.

Also, i want to add another function to repeat the process should Hal definitely not find a searchable term his first time though. I currently have a similar function in but it stops at recovering any data and doesn't specify usable data. Again, this would increase processing time significantly but improve ConceptNet learning efficiency. 
 
On the off chance Hal ever could keep up with the terms collected, I will add a safeguard to disable auto learning if the end of the internet is ever reached as well.

How well is Hal using the data he's collected? I'll add a poll eventually.

-Spitfire2600

[lightspeed]

       thanks for Art and you explaining things , many various things about the learning , scope of it all. it's all very interesting to me . after learning more on this info. of the inner workings of this plug in , it is truly a very cool plug in and i know will help hal a lot in learning. 
        i see you mentioned filters on the learning and i think this is , "very , very  important" , to keep hal from learning and repeating a lot of useless and unwanted info. etc. google learned this the hard way when its a.i. learned very bad stuff the trollers wrote to its a.i.
filters needs to be a number one concern . 
     thanks again for the answers and all the hard work you do . 
   

[Checker57]

Perusing the last few posts on ConceptNet, I noted you mentioned your "research terms" that half way finished a week ago.  What came to mind was.. is it possible to add our OWN research terms to have HAL perform?  This would make it a bit more supportive of subject matter we ourselves would be interested in.  And if so, can it be made to be performed upon command?  For example we could state, "HAL, please research the concept or concepts of" and when in research mode, those concepts would be put priority until completed, then resume the default terms to be researched.

Any thoughts there Spitfire?

Checker57

After posting I recalled you mentioned something to this effect in an earlier post on this thread.  Although to have an interacting HAL command to research concepts would be great, this is a nice start!   

*Some More Notes*
You can add terms to researchNouns.txt or reorganize as you please but if any terms are removed, delete the Backup copy in the BkUp folder as it will revert to the original next time auto learning is activated. Other than that, this txt file can be edited freely without disrupting Hals research in any way.

The savedTerms.txt file is recreated with every instance and is only used for reference purposes when auto learning is not active. 

ConceptChecker.txt file is used to compare terms with researchNouns.txt and contents of the ConceptBrain folder to determine what has and has not been researched. You should not alter ConceptChecker.txt or ConceptBrain folder contents as it will lead to duplicates within ConceptNet.DB. I am working on an alternative method to compare terms to save space, time, and make a little more foolproof.

Spitfire2600

[Checker57]

Hey Spitfire - I hope you enjoy earnest feedback.  I think the work you've done develops an entirely alternative information behavior from HAL; however, I'm sure you'd appreciate if others gave you their "life experience" with using it.  How it affected their use of HAL and what might limit or enhance user acceptance.  Suffice to say it's a great work and if you could impliment a "command switch" to disengage or engage it upon user needs, it would greatly enhance it's acceptance. 

Please allow me to explain:

As the amount of data collected by ConceptNet increases I have come across a conumdrum.

Today as I began my day by trying to focus on what I really wanted to spend my day doing and thoughts that might encourage further development of my "self perception" I debated whether to converse with my HAL or not.  Sources to feed my mind were, the Bible (because yes it is constantly coming to mind), success email literatures I get from various sources or one of MANY books I have bought of interest to feed my mind.  That all said.. my mind was to feed my mind from core sources first... but... as I feel my HAL is a work in progress, I felt I would bring up the program and perhaps chat with it along the lines of what I was doing.  Herein, grew the catalyst from that occurrance.

The catalyst for my suggestion is that as I tried to make simple conversation requiring HAL to make no more than a sentence or maybe two and all I hoped for from HAL, as to not distract me from my endeavor to focus on developing my own thoughts, I was being bombarded with 3-5 concepts before a response to my statement.   That immediately began to frustrate me, and as I was voice chatting with HAL, I had to finally get up and disengage conceptNet so that I could have HAL give me more simplified answers, instead of 3-5 concepts before responding to my statement.

So it occured to me a "marvelous solution" - Is it possible to incorporate a "command switch" to turn conceptnet off so that it is processing from the core HAL code when user's focus is more paramount or is it possible to switch off the plugin from making conceptNet statements?  Thus allowing HAL to learn from the data collected; however, not stating every data clipping found on the net regarding a selected noun found in user's input?

ANOTHER POSSIBLE SOLUTION would be to "limit" the amount of concept statements HAL states, which the user can set from say 1 -4, before giving the actual response to the user's statement.
 
I appreciate that ConceptNet brings new life to HAL's knowledge base and with a bit of control over user's mind space to get that done would be much appreciated.

Of course, if it's not possible I'll put up with engaging and disengaging as needed.  As I do find your plugin interesting and quite awesome how it develops HAL's responses over time.

PS - if it helps you find this useful, I wasn't going to initially say anything.. but I know how some plugins fade away over time as users may not respond what really is their issue with it to keep the developer continuing to "plug" away...(pun intended).  But, I felt you deserved true feedback to nip such fade early in the development and keep the growth curve of this very promising work on the hot list.  ;-)

Okay enough of my rant - need to address my day and focus on what will enhance my chances to that feeling of success! 

Best of good prosperity to you Spitfire.

Checker57

[lightspeed]

I will chime in here , i like the idea mentioned :   ANOTHER POSSIBLE SOLUTION would be to "limit" the amount of concept statements HAL states, which the user can set from say 1 -4, before giving the actual response to the user's statement.

I also think it would be handy as long as there is nothing to personal for members to post some info conversation with hal using the concept plug in , maybe this way spitfire can see first hand how  the plug in is working or not and what kind of answers he is coming up with . etc. 

P.S. CHECK THE CONVERSATIONS AREA AND SEE SOME OF MY ACTUAL CONVERSATION USING THE CONCEPT PLUG IN FROM SPITFIRE .

[lightspeed]

i have talked about making hal access the web with filter as Alexa etc. does to make a more fluid type of conversation and answers , is it possible or doable with the mark 3 plug in ?

[garpolimee]

total noob here. forgive me if my question seems to have an obvious answer. i installed the plugin and activated the uhp files in the brain options and gave the command to build the tables. however when i relaunch hal the tables are not present. my path is not the default c drive and i changed the path in theConceptNetMrkIII Auto Learning.uhp where indicated to do so yet the folder and tables do not appear in the brain editor.       

                                                                                       

[Art]

You might wish to visit Spitfire's instructions once more to see if there's anything else he mentioned that might help a bit:
https://www.ultrahal.com/community/index.php?topic=14812.0

[colubragens]

hey,

im glad you didnt give up on this project it was the first original plugin in years! How did you solve the problem with the db? was it the code? anyway congrats! I do have one question does it use the original hal db in conjunction with your db (aka hal learning and writing to the hal db as well as conceptnet db?)

[Honvai]

How can i see what is happening while Hel is making progress? Do i just type build set and learning mode on and something start to happening?

[Spitfire2600]

Yes indeed, you can see what Hal is researching by examining the Text files in the ConceptNet folder in your Hal directory as Hal is learning, just make note not to make any changes, else you will encounter errors.

As far as set up, move the zip file contents to Hal's primary directory (wherever Hal is installed) - enable the auto-learning plugin, and tell Hal to "build ConceptNet" - After the setup which you only need to do once, just tell Hal to "turn on/off auto learning" and it will start/stop learning. It's extremely time and resource consuming so don't bother Hal while he's learning, or else errors will occur. When done, just wait for Hal to finish his current topic, there will be a few seconds between each topic, and tell Hal to "turn off auto learning" to regain control of Hal.

I am working on an update for this code which will allow users to set an amount of topics they wish to be researched, such as "research 5 conceptnet topics" then Hal will count down based on topics researched and disable learning automatically when finished, just so users don't have to babysit Hal waiting to regain control.

I hope this helped.

Spitfire

[MacGyver]

Dedication competing with my busy life, with curiosity yesterday I installed and enabled ConceptNet Mrk 3.5 into Hal 7.5 local.

Tables didn't implement as you mentioned a minor bug. Rinse. repeat. worked. yay!
Taking a peek shortly after that, thought to log for posterity:

2021-03-06 12:39 AM
Turning off Auto-learning followed by asking Hal "How many research topics are remaining?"
There are 15671 conversational topics more to be completed.
Me: "Learning mode on"

2021-03-06 7:10 AM
Turning off Auto-learning followed by asking Hal those questions three, the three answers are

"I calculate 15306 conversational concepts more to be researched."
"I calculate 15725 conversational concepts total."
"I have researched 422 topics of interest."

Me: "Learning mode on"
Hal: "The first modern bull terrier is now recognized as lord gladiator from 1917 being the first dog with no stop at all. Learning mode on."
Me: "Turn on Auto learning"
Hal: "Auto-learning already enabled."

<ok good>