Brainy Bot Files
Published by Christy Dena January 13th, 2006 in Uncategorized, bots, Software.Here is a list of places that offer (mostly) free data for you to add knowledge to your bot. This can be done through a live feed and or in most cases by downloading the files for extraction from a database. I personally have not utilised these as yet but can see the benefits for particular applications. I’m not for the attempt to create a bot that knows everything, just to be able to know some things…and these databases are a whole lot quicker than entering every possible fact!
AnswerBus
British National Corpus
CIA World Factbook + Factbook on Intelligence
Linguistic Data Consortium
MIT’s Open Mind Commonsense
MIT’s START
Moby Lexicon Project
OpenCyc
Open Mind
Open Mind [Concept Net]
The Consortium for Lexical Research
Thought Treasure
John Bateman’s Generation Bank
John Bateman et.al’s KOMET-Penman Multilingual Linguistic Resource Development Environment (KPML)
WordNet
WordNet [alphabetic version]
WordNet 2.0 in MySQL
I’m interested in others that people recommend, as well as comments on ones they have used.
This is a great resource, Christy. I first encountered Thought Treasure in 1997, although at the time I didn’t realize there were earlier applications (such as Cyc). A lot of interesting (and interrelated) projects are coming out of MIT.
Has anyone had practical experience interfacing one of these systems with a bot setup such as A.L.I.C.E./Pandorabots etc.? It seems to me that either harvesting information from these systems or else building a bridge between them and a bot is always a project in and of itself. For example I’ve heard of CyN:
In a quick search, I also found one research paper on A.L.I.C.E. + CommonSense.
Does anyone know of any similar initiatives as an alternative to by-hand database harvesting?