Newsgroups: comp.ai.fuzzy
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!news.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!swrinde!elroy.jpl.nasa.gov!lll-winken.llnl.gov!ames!waikato!canterbury.ac.nz!otago.ac.nz!is30224
From: is30224@otago.ac.nz
Subject: Estimating effort
Message-ID: <1994Oct7.194915.1@otago.ac.nz>
Lines: 16
Sender: usenet@news.otago.ac.nz (News stuff)
Nntp-Posting-Host: velos.otago.ac.nz
Organization: University of Otago, Dunedin, New Zealand
Date: Fri, 7 Oct 1994 06:49:15 GMT

Hi everyone!
        I am currently carrying out research in estimating software developemnt
effort using fuzzy systems.  I am using as inputs specification-based
complexity measures, such as total number of entities in data model and total
number of screens displayed.  The outputs I am trying to get are measures of
the development effort, such as analysis-design time and total time.
	I have had some success with small fuzzy expert systems, using both
intuitive rules and rules extracted from a trained neural network.  Any advice
as to alternative techniques would be welcomed.  I am also interested in any
alternative ways of extracting information for fuzzy systems from trained
networks.
	Reply to me and I will prepare a summary for the group.

Thanks in advance
Andrew Gray
	
