User:Tancre/Prototyping/AI/ELIZA: Difference between revisions

From XPUB & Lens-Based wiki
No edit summary
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
= ELIZA =
= ELIZA =
[[File: srycntrmbr.png | frame]]
<br><br><br><br><br>
[http://pzwart1.wdka.hro.nl/~mmurtaugh/eliza2018/eliza.html ELIZA]
<br><br><br><br><br>
== Process ==
>> input sentence <br>
1. The sentence is broken into words <br>
2. Set of pre-substitutions [pre:] <br>
3. List of recognized keywords + sorting on priority [key:] <br>
4. Research of a list of decomposition patterns. The first that matches is selected + list of sinonyms [decomp: + synon:] <br>
5. Selection of a reassembly pattern for the selected decomposition pattern. If already selected goes on the next until it restarts. [reasemb:] <br>
6. Set of post-substitutions [post:] <br>
>> output sentence


== Process ==
== Suggested Implementations ==
: > input sentence
* augmentations of its knowledge
# Pattern of decomposition of the sentence
** access to a store of informations
# Set of pre-substitutions
** from its conversational experience
# List of recognized keywords + sorting on priority
* built a model of the subject conversing with it (belief structure)
# Research of a list of decomposition patterns. The first that matches is selected
 
# Selection of a reassembly pattern for the selected decomposition pattern. If already selected goes on the next until it restarts.  
== Links ==
# Set of post-substitutions
[http://web.stanford.edu/class/cs124/p36-weizenabaum.pdf 1966 Weizenbaum - ELIZA, a computer program for the study of natural language communication between man and machine] <br>
: > output sentence
[https://www.atariarchives.org/bigcomputergames/showpage.php?page=20 1984 BASIC ELIZA in big computer games]<br>
[http://web.stanford.edu/group/SHR/4-2/text/dialogues.html 1995 Güzeldere, Franchi - Dialogues with colorful personalities of early AI]<br>
[http://elizagen.org/index.html elizagen.org]<br>
[https://en.wikipedia.org/wiki/ELIZA wikipedia]<br>
[https://www.youtube.com/watch?v=Rk7QQfifov4 ELIZA VS ELIZA]

Latest revision as of 00:38, 22 October 2018

ELIZA

Srycntrmbr.png






ELIZA




Process

>> input sentence
1. The sentence is broken into words
2. Set of pre-substitutions [pre:]
3. List of recognized keywords + sorting on priority [key:]
4. Research of a list of decomposition patterns. The first that matches is selected + list of sinonyms [decomp: + synon:]
5. Selection of a reassembly pattern for the selected decomposition pattern. If already selected goes on the next until it restarts. [reasemb:]
6. Set of post-substitutions [post:]
>> output sentence

Suggested Implementations

  • augmentations of its knowledge
    • access to a store of informations
    • from its conversational experience
  • built a model of the subject conversing with it (belief structure)

Links

1966 Weizenbaum - ELIZA, a computer program for the study of natural language communication between man and machine
1984 BASIC ELIZA in big computer games
1995 Güzeldere, Franchi - Dialogues with colorful personalities of early AI
elizagen.org
wikipedia
ELIZA VS ELIZA