TY - GEN

T1 - Python-based fuzzy classifier for cashew kernels

AU - Tomar, Snehal Singh

AU - Narendra, V. G.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Fuzzy logic is a well-known branch of mathematics which provides a quantitative framework to discuss uncertain events and hence make logical estimations for uncertain outcomes. In this work, the key objective is to explore and illustrate the tools and techniques required to perform fuzzy operations and hence realize a basic fuzzy classifier in Python and assert its applicability over other conventional fuzzy logic tools such as the fuzzy logic toolbox in MATLAB. The above-mentioned classifier took real-world data of physical parameters such as length, width and thickness of white wholes cashew kernels which had highly overlapping data ranges as input and classified them into suitable categories. The observed computation time for successful (crisp) classification of the kernels into WW-320, WW-240, WW-210 and WW-180 categories using the said classifier was 0.43, 0.43, 0.42 and 0.46 s, respectively, whereas the fuzzy logic toolbox in MATLAB took minimum 0.58 s only to obtain a fuzzy output on the same computing system.

AB - Fuzzy logic is a well-known branch of mathematics which provides a quantitative framework to discuss uncertain events and hence make logical estimations for uncertain outcomes. In this work, the key objective is to explore and illustrate the tools and techniques required to perform fuzzy operations and hence realize a basic fuzzy classifier in Python and assert its applicability over other conventional fuzzy logic tools such as the fuzzy logic toolbox in MATLAB. The above-mentioned classifier took real-world data of physical parameters such as length, width and thickness of white wholes cashew kernels which had highly overlapping data ranges as input and classified them into suitable categories. The observed computation time for successful (crisp) classification of the kernels into WW-320, WW-240, WW-210 and WW-180 categories using the said classifier was 0.43, 0.43, 0.42 and 0.46 s, respectively, whereas the fuzzy logic toolbox in MATLAB took minimum 0.58 s only to obtain a fuzzy output on the same computing system.

UR - http://www.scopus.com/inward/record.url?scp=85058993230&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058993230&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-1592-3_28

DO - 10.1007/978-981-13-1592-3_28

M3 - Conference contribution

AN - SCOPUS:85058993230

SN - 9789811315916

T3 - Advances in Intelligent Systems and Computing

SP - 365

EP - 374

BT - Soft Computing for Problem Solving - SocProS 2017

A2 - Bansal, Jagdish Chand

A2 - Nagar, Atulya

A2 - Ojha, Akshay Kumar

A2 - Das, Kedar Nath

A2 - Deep, Kusum

PB - Springer Verlag

T2 - 7th International Conference on Soft Computing for Problem Solving, SocProS 2017

Y2 - 23 December 2017 through 24 December 2017

ER -