Nautilus Systems, Inc. logo and menu bar Site Index Home
News Books
Button Bar Menu- Choices also at bottom of page About Nautilus Services Partners Case Studies Contact Us
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Subscribe]

DM: FUZZYMANAGER software released

From: Andrew Gray
Date: Thu, 29 Jul 1999 09:59:35 -0400 (EDT)
  • Organization: University of Otago


We have just released the first version of the FUZZYMANAGER 
software package for Microsoft Windows (95/NT, should run under 
98 but this hasn't yet been tested).  This was developed within the 
Software Metrics Research Laboratory (SMRL) here at the 
University of Otago.  The software currently consists of two 
modules called FULSOME (Fuzzy Logic for Software Metrics) and 
CLUESOME (Cluster Extraction for Software Metrics).   

As the names suggest the software was originally developed with 
the requirements of project managers/software metricians in mind.  
Thus, the software is intended to be very easy to use, guides the 
user through the steps of creating fuzzy systems, has a very 
graphical interface to explain the inference process, has a variety 
options for experimenting, etc.   

However, despite the name, as this list of features suggests it 
could also be used just as well by anyone wanting a simple 
interface to fuzzy logic systems and fuzzy clustering.  It will be 
used for an introductory course segment on fuzzy logic and we 
think that it has some promise for teaching purposes.  

The FULSOME software allows for data editing, membership 
creation, rule editing, and inference with a reasonable variety of t- 
norms, t-conorms, defuzzication methods, and system options.   

CLUESOME uses the same file formats to extract membership 
functions and/or rules from data (and existing membership 
functions if only rules are desired) based on fuzzy-clustering 
algorithms.  This can, optionally, all be accomplished using a 
wizard interface.  These systems are generally adequate as rough 
drafts for an expert to refine, but perform quite well when compared 
to other techniques such as regression (where the generalisation 

performance is much better based on the results from simulations 
using various small data sets) even when used without expert 

The software can be downloaded from the below web page.  

This is part of our new web site which is currently being built up 
from our old site.  As such most of the pages are still blank, but 
the existing web site below is still available.  

More recent papers (including a few on FUZZYMANAGER that 
were published at recent conferences), presentations, etc. than 
those that are on the old server will eventually appear on the new 

This is the first release of the software so we would be especially 
grateful for suggestions, criticisms, and any offers of collaboration 
with further development (testing, success stories, etc.).  We have 
several other modules that are under development that will 
hopefully be released soon.   

We are hoping to make new releases of FUZZYMANAGER every 
couple of weeks with changes made based on the feedback we 
receive.  So please send your comments as soon as possible.  

We also provide four listserv-based mailing lists for announcing new 
releases of FUZZYMANAGER, discussing FUZZYMANAGER, 
fuzzy logic for project management, and computational intelligence 
techniques for software engineering that may be of interest to 
anyone downloading the software (especially the first two).   


Software Metrics Research Laboratory
Department of Information Science, University of Otago
phone +64 3 479 5282 fax +64 3 479 8311

[ Home | About Nautilus | Case Studies | Partners | Contact Nautilus ]
[ Subscribe to Lists | Recommended Books ]

logo Copyright © 1999 Nautilus Systems, Inc. All Rights Reserved.
Mail converted by MHonArc 2.2.0