A Python Software Library for Computing with Words and Perceptions

Computing with Words (CWW) methodology has been used to design intelligent systems which make decisions by manipulating the linguistic information, like human beings. Human beings naturally understand (and express) themselves linguistically, and hence can reason (and make decision) just with linguis...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) pp. 1 - 6
Main Authors Sharma, Deepak, Gupta, Prashant K., Andreu-Perez, Javier, Mendel, Jerry M., Lopez, Luis Martinez
Format Conference Proceeding
LanguageEnglish
Published IEEE 11.07.2021
Subjects
Computing with Words
Fuzzy sets
Intelligent systems
Interval Approach
Linguistics
Numerical models
Perceptual Computing
Python toolbox
Semantics
Software libraries
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:Computing with Words (CWW) methodology has been used to design intelligent systems which make decisions by manipulating the linguistic information, like human beings. Human beings naturally understand (and express) themselves linguistically, and hence can reason (and make decision) just with linguistic information without any numerical measure. Perceptual Computing makes use of type 2 fuzzy sets for modeling the words in the CWW paradigm. This use of type-2 fuzzy sets enables better representation of the inherent uncertainty in the fuzzy linguistic semantics on numerous problems. To realise the potential of Perceptual Computing, its MATLAB implementation has been made freely available to the end-users/ researchers, and MATLAB is a proprietary development environment. Therefore, this contribution aims at proposing a python implementation of the Perceptual Computing, or its main processing element the perceptual computer that consists of three components viz., encoder, CWW engine and decoder. Our python implementation provides the end user with a seamless blending amongst all three components, which does not exist yet, to the best of our knowledge.
ISSN:1558-4739
DOI:10.1109/FUZZ45933.2021.9494557