Fuzzy inference system model from non-fuzzy clustering output
Fuzzy Inference System (FIS) is a process of mapping input into the desired output using fuzzy logic theory where decisions can be made or patterns are discerned. This study aims to discuss on how non-fuzzy clustering output can be used to construct a model of FIS. Here, the proposed idea is to show...
Saved in:
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Medwell Publications
2019
|
Subjects: | |
Online Access: | http://eprints.uthm.edu.my/4076/1/AJ%202019%20%28215%29.pdf http://eprints.uthm.edu.my/4076/ https://dx.doi.org/ 10.36478/jeasci.2019.4035.4042 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Fuzzy Inference System (FIS) is a process of mapping input into the desired output using fuzzy logic theory where decisions can be made or patterns are discerned. This study aims to discuss on how non-fuzzy clustering output can be used to construct a model of FIS. Here, the proposed idea is to show the efficient use of the FIS as a prediction model for the data classification. In this study, employment income, self-employment income, property and transfer received are taken into account for clustering the household income data. Then, the FIS prediction model is built using the center values of clusters formed and the output of FIS is compared to the original cluster in which the best fit prediction model to the data is determined. In conclusion, the best prediction model in identifying income class is discovered based on the Root Mean Square Error (RMSE) value computed. |
---|