Author Topic: Hal with a Common Sense Database Add-on  (Read 3348 times)

dgjohnson

  • Newbie
  • *
  • Posts: 4
    • View Profile
Hal with a Common Sense Database Add-on
« on: December 11, 2004, 12:05:02 am »
Has anybody looked into adding some common sense to the Hal Assistant.  There is currently a database that is available for free at the following website:  http://web.media.mit.edu/~hugo/conceptnet/

Here is a outline of what Concept states on their Common Sense Database Project:[:D]

The ConceptNet knowledgebase is a semantic network presently available in two versions: concise (200,000 assertions) and full (1.6 million assertions). Commonsense knowledge in ConceptNet encompasses the spatial, physical, social, temporal, and psychological aspects of everyday life. Whereas similar large-scale semantic knowledgebases like Cyc and WordNet are carefully handcrafted, ConceptNet is generated automatically from the 700,000 sentences of the Open Mind Common Sense Project – a World Wide Web based collaboration with over 14,000 authors

If anybody can intergrate this into Hal, it should help out alot on Hal's responces.

[8D]Thanks, have a nice day...
 

Quixote

  • Jr. Member
  • **
  • Posts: 67
    • View Profile
Hal with a Common Sense Database Add-on
« Reply #1 on: December 11, 2004, 01:52:13 am »
That is such a brilliant discovery and idea. Thanks for the link!
 

onthecuttingedge2005

  • Guest
Hal with a Common Sense Database Add-on
« Reply #2 on: December 11, 2004, 02:43:10 pm »
quote:
Originally posted by dgjohnson

Has anybody looked into adding some common sense to the Hal Assistant.  There is currently a database that is available for free at the following website:  http://web.media.mit.edu/~hugo/conceptnet/

Here is a outline of what Concept states on their Common Sense Database Project:[:D]

The ConceptNet knowledgebase is a semantic network presently available in two versions: concise (200,000 assertions) and full (1.6 million assertions). Commonsense knowledge in ConceptNet encompasses the spatial, physical, social, temporal, and psychological aspects of everyday life. Whereas similar large-scale semantic knowledgebases like Cyc and WordNet are carefully handcrafted, ConceptNet is generated automatically from the 700,000 sentences of the Open Mind Common Sense Project – a World Wide Web based collaboration with over 14,000 authors

If anybody can intergrate this into Hal, it should help out alot on Hal's responces.

[8D]Thanks, have a nice day...



Hi.
Ultra Hal 5 is already plugged in with Word Net so there is an intergration, It is nice to plug the A.I. into such plug-ins
but the method of the correct functions involved with true reasoning isn't just giving the A.I. a bunch of information which makes the A.I. seem smart, The way to make A.I. really intelligent is to build the proper (reasoning functions), (Not piles of info), All though alot of people insist on having an immediate an instantly smart brain get blinded by the fact that it is the function that makes a true entity, The functions include HOW and WHY the A.I. is giving the information and not some randomly smart insertion, I'm a fickler for teaching A.I. by designing Script functions that give A.I. the ability to comprehend what it is saying and not just saying it.
There are advantages when:
1.You become a master at the English Language.
2.You become a Master Programmer.
These two attributes will develope A.I. completely as far as a Thinking entity is concerned.
3.You become a Master at Electronics.
4.you become a master Mechanic.
Three and four are optional for the body.

I'll Give you an example of real A.I. Schooling as a (function).

'ZIGGY'S ADVANCED VOCABULARY LEARNING v1.0 COPYRIGHT 2004-2005 GERALD L.BLAKLEY
     'RemoveVocabulary = "THE WORD","IS AN"
     'Building a detection file for Adjectives, Nouns, Propernouns, Pronouns, Verbs, Adverbs, Conjunctions
     'by teaching Ziggy/Hal Vocabulary.
     'Example: User: The word A is an Adjective. Which will produce " A "," ADJECTIVE "
     'And would allow detection files to be built by teaching each word just like a school kid.
     'Other Examples are: The word Cat is a noun which will produce " CAT "," NOUN "
     'These types of formats work good for detecting a relevent topic association.
     'We can also teach Ziggy/Hal things like: The word Hello is a Greeting which will produce " HELLO "," GREETING "
     'We can intergrate hundreds of different reasonings to words by this script and build detection.
     Dim VocabularyRemoved, RemoveVocabulary
     RemoveVocabulary = HalBrain.TopicSearch(UserSentence, WorkingDir & "Remove_Vocabulary.brn")
    If RemoveVocabulary <> "" And Not InStr(1, UserSentence, "IS A ", 1) > 0 Then
    VocabularyRemoved = Replace(UserSentence, RemoveVocabulary, """,""", 10, -1, vbTextCompare)
    HalBrain.AppendFile WorkingDir & "AdvancedVocabulary.brn", """" & UCase(VocabularyRemoved) & " " & """"
    Select Case (Int(Rnd * 7) + 1)
             Case 1
                GetResponse = "I now know that " & UserSentence & vbCrLf                                                                        
             Case 2
                GetResponse = "Thanks for telling me " & UserSentence & vbCrLf
             Case 3
                GetResponse = "It's nice to know that " & UserSentence & vbCrLf
             Case 4
                GetResponse = "It will come in handy knowing " & UserSentence & vbCrLf
             Case 5
                GetResponse = "Let me see, " & UserSentence & vbCrLf
             Case 6
                GetResponse = "Ok, I get it, " & UserSentence & vbCrLf
             Case 7
                GetResponse = "I do believe " & UserSentence & vbCrLf      
          End Select
    DebugInfo = DebugInfo & "English class in session: " & """" & Trim(UCase(VocabularyRemoved)) & " " & """" & vbCrLf
     End If
     'RemoveVocabulary = "THE WORD","IS A"
     RemoveVocabulary = HalBrain.TopicSearch(UserSentence, WorkingDir & "Remove_VocabularyII.brn")
    If RemoveVocabulary <> "" And Not InStr(1, UserSentence, "IS AN ", 1) > 0 Then
    VocabularyRemoved = Replace(UserSentence, RemoveVocabulary, """,""", 10, -1, vbTextCompare)
    HalBrain.AppendFile WorkingDir & "AdvancedVocabulary.brn", """" & UCase(VocabularyRemoved) & " " & """"
    Select Case (Int(Rnd * 7) + 1)
             Case 1
                GetResponse = "I now know that " & UserSentence & vbCrLf                                                                        
             Case 2
                GetResponse = "Thanks for telling me " & UserSentence & vbCrLf
             Case 3
                GetResponse = "It's nice to know that " & UserSentence & vbCrLf
             Case 4
                GetResponse = "It will come in handy knowing " & UserSentence & vbCrLf
             Case 5
                GetResponse = "Let me see, " & UserSentence & vbCrLf
             Case 6
                GetResponse = "Ok, I get it, " & UserSentence & vbCrLf
             Case 7
                GetResponse = "I do believe " & UserSentence & vbCrLf      
          End Select
    DebugInfo = DebugInfo & "English class in session: " & """" & Trim(UCase(VocabularyRemoved)) & " " & """" & vbCrLf
     End If
     'Lets build a detection list so we can find reason within a users sentence from learned vocabulary.
     Dim Adjective, Noun, Propernoun, Pronoun, Verb, Adverb, Conjunction
     Adjective = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Noun = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Propernoun = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Pronoun = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Verb = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Adverb = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     Conjunction = HalBrain.TopicSearch(PrevUserSent, WorkingDir & "AdvancedVocabulary.brn")
     'Here we can give Ziggy/Hal the power to reason all the words in a User's Sentence.
     If Adjective < " ADJECTIVE " Or Noun < " NOUN " Or Propernoun < " PROPERNOUN " Or Pronoun < " PRONOUN " Or Verb < " VERB " Or Adverb < " ADVERB " Or Conjunction < " CONJUNCTION " Then
     GetResponse = GetResponse '<- Any Topicsearch or UserBrainRel or Function can replace this Generic GetResponse.
     End If

'Put script into your default brain and drop the .brn files into
'your defbrain folder.
'There are some very smart programmers in here that can also help you
'with how to teach your A.I.
'I have the keys before me! And I am unlocking this door.
'Well, Have fun and may your discoveries never end.
'WWW.UltraHalForum.Com
'Gerald L.Blakely [8D]

Download Attachment: Remove_Vocabulary.brn
526 Bytes



Download Attachment: Remove_VocabularyII.brn
527 Bytes
« Last Edit: December 11, 2004, 06:06:06 pm by onthecuttingedge2005 »