From timm@cse.unsw.edu.au Tue Sep  3 21:01:21 1996
Date: Tue, 3 Sep 1996 16:11:20 +1000 (EST)
From: Tim Menzies <timm@cse.unsw.edu.au>
To: Vasilios Tourloupis <vasilios@insect.sd.monash.edu.au>
Subject: Re: Abstract

Vasilios Tourloupis writes:
> 
> What follows is my abstract:
> 
> In ``Principles for Generalized Testing of Knowledge Bases'', Menzies proposes
> an abductive inference engine, called HT4, which emulates validation.  He argues
---------------------------------------------------^^^^^^^^ implements

> that HT4 can be applied to many knowledge level tasks, and that it can replace,
> rather than augment existing knowledge base system architectures.  He further
> argues that HT4 is practical for most discrete, propositional theory sizes that
> are seen in practice, and concludes that the practical limits to this inference
> engine are the limits to testing, and therefore the limits to model construction
> ~\cite{MENZIESPHD}.
> 
> This has two consequences:
> 
> \begin{enumerate}
> \item Knowledge Base Systems (KBS) be designed based on a large number of
> smaller, totally independent concepts, rather than exploring the inner
> complexities of some particular concept.
> 
> \item If models are to be designed and built larger than the practical limits
> to testing, then an alternative plan should be developed to cater for
> inappropriate behaviour that may arise when the model fails ~\cite{MENZIESPHD}.
> \end{enumerate}
> 
> For these reasons, this research is aimed at optimising the general abductive
> inference engine proposed by Menzies, for efficiency, by applying ``clever''
> algorithmic optimisations proposed in the literature.  We find that out of
> the two methods proposed, only one had limited.  We also find that refining
------------------------------------^^^^^^^^^^^^ worked.
> the requirements of the HT4 algorithm, has achieved a greater success of
> optimisation, with little loss to the generality of the algorithm.  Although
> measuring the limits to testing is beyond the scope of this thesis, we will
> argue that we have increased the limits to testing by increasing the
> efficiency of HT4.
> 
> Regards,
> 
> Bill Tourloupis
> P.S. I will email my conclusion some time later today.
> +-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
> | Vasilios E. Tourloupis             | vasilios@insect.sd.monash.edu.au |
> | Dept. of Software Development      | vasilios@hestia.sd.monash.edu.au |
> | Monash University,                 |                /\                |
> | Caulfield East, Vic., 3145         |___________/\  /  \_______________|
> | Australia                          |             \/                   | 
> +-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
> | Disclaimer: There are some that call me Bill!  But I don't know why?  |
> +-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
> 


-- 
        Dr. Tim Menzies     rm EE339  |  
        timm@cse.unsw.edu.au          |  Hydrogen is a colorless,
 ,-_|\  www.cse.unsw.edu.au/~timm     | odorless gas which, given
/     \ AI Dept, School of Computer   | enough time, turns into
\_,-._* Science &Engineering, Uni NSW |  people.  --Henry Hiebert 
     v  Sydney, Australia, 2052       | 
        +61-2-93854034(p)93855995(f)  |