IFCRL: Interval-Intent Fuzzy Concept Re-Cognition Learning Model

The fuzzy concept serves as a crucial tool for describing phenomena and constitutes the fundamental unit of human cognition. Fuzzy concepts are characterized by their extent and intent, with the latter being comprised of continuous membership degrees. Given that human cognition often progresses from...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on fuzzy systems Vol. 32; no. 6; pp. 3581 - 3593
Main Authors: Ding, Yi, Xu, Weihua, Ding, Weiping, Qian, Yuhua
Format: Journal Article
Language:English
Published: New York IEEE 01.06.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Classification
Classification algorithms
Clustering
Cognition
Cognition & reasoning
Computational modeling
Concept clustering
concept-cognitive learning
Context modeling
Fuzzy systems
Granular computing
interval-intent
Learning
Mathematical models
object classification
Stochastic processes
ISSN:1063-6706, 1941-0034
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The fuzzy concept serves as a crucial tool for describing phenomena and constitutes the fundamental unit of human cognition. Fuzzy concepts are characterized by their extent and intent, with the latter being comprised of continuous membership degrees. Given that human cognition often progresses from vagueness to precision, it is imperative that the form of intent not be confined to a singular continuous value; rather, an interval possesses superior flexibility in this regard. Initial cognitive processes lack comprehensiveness in acquiring knowledge, necessitating subsequent cognitions to more accurately delineate the intended scope of a concept. Motivated by this insight, we proposed an interval-intent fuzzy concept re-cognition learning model (IFCRL). First, this model transforms fuzzy concept intent from a single continuous value into an interval-based representation, which describes the range of attribute values for all objects within the given interval. Second, in order to simulate secondary cognitive processes akin to those exhibited by humans toward phenomena, we present a concept re-cognition learning method capable of effectively scaling intervals within reasonable bounds. Third, aiming to overcome cognitive barriers and emulate imaginative processes observed in human brains, we introduce a concept clustering approach based on intent similarity which significantly reduces concept complexity while enhancing cognitive efficiency. Finally, we evaluate our classification performance using 12 datasets and experimental results demonstrate that IFCRL outperforms 14 other classification algorithms both feasibly and effectively.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1063-6706
1941-0034
DOI:10.1109/TFUZZ.2024.3376569