From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: **
X-Spam-Status: No, score=2.1 required=5.0 tests=BAYES_05,INVALID_DATE,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,79d7c4b0173e5e6a,start
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 1994-12-12 06:35:06 PST
Path: 
 bga.com!news.sprintlink.net!howland.reston.ans.net!spool.mu.edu!cass.ma02.bull.com!scn.ORG!mentifex
From: mentifex@scn.ORG (SCN User)
Newsgroups: comp.lang.ada
Subject: Mind.rexx AI public domain
Date: 12 Dec 1994 14:31:05 -0000
Organization: Bull Worldwide Information Systems.
Sender: daemon@cass.ma02.bull.com
Distribution: world
Message-ID: <199412121430.GAA14280@scn.org>
Reply-To: mentifex@scn.org
NNTP-Posting-Host: cass.ma02.bull.com
Date: 1994-12-12T14:31:05+00:00
List-Id: <comp.lang.ada>

    The Mind.rexx AI source code is free for porting to Ada.

Mindmap30NOV94 is a flowchart of function-calls for Mind.rexx AI.

        ___________    _________________________________________
       |Audition:  |  |Mind.rexx Level Three:  Auditory Memory. |
Input  |User enters|  |ShortTermMem() enters quasi-phonemes     |
>>>>>>>|sentences. |  |(letters) into the engram array "ph.t".  |
       |___________|  |_________________________________________|
         |              /|\
 ________V___          __|_____________          _______________
|Sensorium() |        |Comparator()    |        |String_effect()|
|orchestrates|        |matches input   |        |passes         |
|recognition |------->|phonemes against|------->|activation from|
|of words.   |        |auditory memory.|        |one phonemic   |
|            |        |________________|        |engram to its  |
|            |                                  |successor.     |
|            |       _____________________      |_______________|
|            |      |Main body DO-FOREVER |    _________________
|            |<-----|loop calls functions |   |Metempsychosis() |
|            |      |in simulation of     |   |(not yet coded)  |
|            |----->|massive parallelism. |   |will reload saved|
|            |      |_____________________|   |states of mind.  |
|            |                                |_________________|
|            |       _____________________________________
|            |      |Language() synthesizes a sentence by |Output
|            |      |calling modules that flush out words.|>>>>>>
|            |      |_____________________________________|
|            |            |              |    |
|            |         ___V____________  |  __V_________________
|            |        |Subject() makes | | |Predicate() flushes |
|            |        |a first call to | | |out the most active |
|            |        |NounPhrase().   | | |verb in the lexicon.|
|            |        |________________| | |____________________|
|            |          |                |    |   |    |   |  |
|            |          |   _____________V____V_  |    |   |  |
|            |          |  |Deep_structure()    | |    |   |  |
|            |          |  |transfers activation| |    |   |  |
|            |          |  |from deep concepts  | |    |   |  |
|            |          |  |to shallow lexicon. | |    |   |  |
|            |          |  |____________________| |    |   |  |
|            |         _V_________________________V_   |   |  |
|            |        |NounPhrase() flushes out the |  |   |  |
|            |        |momentarily most active noun |  |   |  |
|            |        |to be subject or object of a |  |   |  |
|            |        |verb around which a sentence |  |   |  |
|            |        |is forming in the robot mind.|  |   |  |
|            |        |_____________________________|  |   |  |
|            |           |    |    |     |             |   |  |
|            |           |    |    |  ___V_____________V_  |  |
|            |           |    |    | |Ideadamp() de-     | |  |
|            |           |    |    | |activates concepts.| |  |
|            |           |    |    | |___________________| |  |
|            |           |    |    |                       |  |
|            |           |    |   _V_______________________V  |
|            |           |    |  |Fiberdamp() deactivates   | |
|            |           |    |  |words that have been used.| |
|            |           |    |  |__________________________| |
|            |           |   _V_____________________________  |
|            |           |  |TransferAct() passes activation| |
|            |           |  |from a temporal node on one    | |
|            |           |  |concept to a synchronous node  | |
|            |           |  |on an associated concept.      | |
|            |           |  |_______________________________| |
|            |         __V____________________________________V_
|            |<-------|Reentry() submits the output of the mind |
|            |        |back into the mind as auditory input.    |
|            |        |It is how a brain "hears itself think."  |
|            |        |_________________________________________|
|            |
|            |     ______________________________________
|            |    |Wordend() is a waystation on the path |
|            |--->|of dealing with recognized words.     |
|____________|    |______________________________________|
     |                                          |
 ____V___________                               |
|Retro() assigns |                              |
|"ultimate-tags" |                              |
|at end of words.|  __________________          |
|________________| |Level Two of Mind:|         |
    |              |Attach() creates  |         |
    |              |quasi-fibers of   |         |
 ___V_________     |lexical items in  |         |
|Newconcept() |    |the compound-stem |      ___V_______________
|records new  |    |array "fiber.n.t",|     |Oldconcept() renews|
|lexical items|--->|which holds the   |<----|old lexical items  |
|and their    |    |controlling       |     |and their          |
|corresponding|    |flag-panel for    |     |corresponding      |
|concepts in  |    |a word stored in  |     |concepts in the    |
|the mindcore.|    |auditory memory.  |     |monad.f.t mindcore.|
|_____________|    |__________________|     |___________________|
       |                                        |      |
       |                                        |      |
 ______V________________________________________V__   _V________
|    Level One:  The Mindcore                      | |Holodyne()|
|Instantiate() holds 64 primitive concepts         | |activates |
|optimized for porting to 64-bit DEC Alpha         | |all nodes |
|and any new concepts learned by Mind.rexx         | |of one    |
|in the compound-stem array "monad.f.t".           | |concept.  |
|__________________________________________________| |__________|

--
Mind.rexx (Amiga) AI is on ftp.wustl.edu in pub/aminet/util/rexx.
Mind.anim and theory of Mind.rexx are on Amiga Library Disk #411.
Theory is on CD-UniBase (tm) from CD Eureka Inc. at 818-789-0269.
You may download latest Mind.rexx from Amiga BBS at 206-744-1254.