Related Experiment Video
Updated: Sep 3, 2025
08:05
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
10.8K
An Explainable Evolving Fuzzy Neural Network to Predict the k Barriers for Intrusion Detection Using a Wireless
Paulo Vitor de Campos Souza1, Edwin Lughofer1, Huoston Rodrigues Batista2
1Institute for Mathematical Methods in Medicine and Data Based Modeling, Johannes Kepler University Linz, 4040 Linz, Austria.
Sensors (Basel, Switzerland)
|July 27, 2022
Summary
This study introduces an interpretable evolving fuzzy neural network for data stream regression. The novel method effectively predicts wireless sensor data for security, outperforming existing approaches with lower error rates.
Area of Science:
- Artificial Intelligence
- Machine Learning
- Data Science
Background:
- Evolving fuzzy neural networks offer adaptive solutions for complex problems by extracting knowledge from data.
- Interpretability is crucial for understanding the behavior of analyzed problems and validating extracted knowledge.
Purpose of the Study:
- To apply an evolving fuzzy neural network for data stream regression with enhanced interpretability.
- To address the challenge of predicting wireless sensor data for security applications, specifically identifying unauthorized entry through k-barrier prediction.
Main Methods:
- Utilized an evolving fuzzy neural network model for data stream regression.
- Implemented an interleaved-predict-and-then-update procedure for evaluating model performance in a streaming context.
- Compared the proposed method against state-of-the-art evolving techniques.
Main Results:
- Achieved significantly lower Root Mean Square Error (RMSE) values on separate test datasets compared to existing methods.
- Demonstrated lower accumulated Mean Absolute Errors (MAEs) in the stream-based evaluation.
- Generated interpretable fuzzy rules, enabling an explainable evaluation of the regression problems.
Conclusions:
- The proposed evolving fuzzy neural network provides a highly interpretable and effective solution for data stream regression.
- The method shows superior performance in predicting wireless sensor data for security, offering practical insights through explainable fuzzy rules.
