IFSMetrics is a package containing modules whereby you can build intuitionistic fuzzy sets (IFSs) from real-world data, perform similarity comparisons between those IFSs and generate a comprehensive report with the results of such comparisons. As such, IFSMetrics aims for the design of similarity measures that perform reliable comparisons between IFSs.
The current version of IFSMetrics consists of three modules: IFSBuilder, IFSComparer and IFSSimReporter. Since these modules are written in C#, you can develop them on platforms like Linux, Windows and macOS where the .Net framework is supported.
IFSMetrics uses the following external components:
| IFSMetrics Module | External Component(s) |
|---|---|
| IFSBuilder (Building process) | svmlight |
| IFSSimReporter (Reporting process) | gnuplot, pdflatex |
In the current version of IFSMetrics, IFSs are derived from the RCV1-v2 dataset:
- Lewis, D. D. RCV1-v2/LYRL2004: The LYRL2004 Distribution of the RCV1-v2 Text Categorization Test Collection (5-Mar-2015 Version). http://www.jmlr.org/papers/volume5/lewis04a/lyrl2004_rcv1v2_README.htm.
The getting started wiki page explains how to start using IFSMetrics.
IFSMetrics is released under the Apache License, Version 2.0.
- IFSBuilder Tutorials
- IFSComparer Tutorials
- IFSSimReporter Tutorials