PPubMed10 Jan 2023Cited 14×

An RBF neural network based on improved black widow optimization algorithm for classification and regression problems

Hui Liu, Guo Zhou, Yongquan Zhou, Huajuan Huang, Xiuxi Wei

Abstract

Introduction Regression and classification are two of the most fundamental and significant areas of machine learning. Methods In this paper, a radial basis function neural network (RBFNN) based on an improved black widow optimization algorithm (IBWO) has been developed, which is called the IBWO-RBF model. In order to enhance the generalization ability of the IBWO-RBF neural network, the algorithm is designed with nonlinear time-varying inertia weight. Discussion Several classification and regression problems are utilized to verify the performance of the IBWO-RBF model. In the first stage, the proposed model is applied to UCI dataset classification, nonlinear function approximation, and nonlinear system identification; in the second stage, the model solves the practical problem of power load prediction. Results Compared with other existing models, the experiments show that the proposed IBWO-RBF model achieves both accuracy and parsimony in various classification and regression problems.

A figure from An RBF neural network based on improved black widow optimization algorithm for classification and regression problems — fig. from the paper

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

The IBWO‑RBF model achieves high accuracy and compact networks across classification, regression, and power load prediction tasks.

What’s new

i.Development of the IBWO‑RBF model that integrates an improved black‑widow optimization algorithm with radial basis function neural networks
ii.Design of a nonlinear time‑varying inertia weight to enhance the algorithm's generalization ability
iii.Comprehensive evaluation on UCI classification datasets, nonlinear regression problems, and short‑term power load forecasting
iv.Demonstration of higher accuracy with fewer hidden neurons compared to 12 competing algorithms

How they did it

The authors designed the IBWO‑RBF model by coupling the improved black‑widow optimization (IBWO) algorithm—augmented with a nonlinear time‑varying inertia weight—to train the parameters of an RBF neural network. Experiments used 10 binary UCI classification datasets (population size = 50, 250 iterations, 10 independent runs) and several regression tasks, including power load prediction. Performance was compared against twelve existing metaheuristic and gradient‑based algorithms.

What they found

Achieved 81.75% classification accuracy on the Blood UCI dataset, ranking second only to GB‑RBF (81.18%) and outperforming other algorithms
Recorded the lowest RMSE of 357.8041 in short‑term power load forecasting, better than BWO‑RBF (565.96) and other single models
Obtained the best MSE of –33.9305 dB in nonlinear function approximation, surpassing BWO‑RBF (–4.5217 dB) and kernel‑based RBF variants
Used fewer hidden neurons than competing methods on most classification datasets, e.g., only 3 neurons on the Blood dataset versus up to 10 for others

in plain words

The researchers created a new kind of neural network that learns using a clever optimization trick called the improved black‑widow algorithm. This makes the network both accurate and small, so it works well on tasks like recognizing patterns and predicting electricity use. They tested it on many standard data sets and real‑world power‑load data and it performed better than many other methods.

why it matters

The IBWO‑RBF approach provides more accurate and parsimonious models for both classification and regression, enabling safer operation and better planning of power grids through improved short‑term load forecasts.

Follow the thread

improved black widow optimization algorithm radial basis function neural networks classification and regression power load prediction metaheuristic

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Similar papers in all four archives

The same idea searched in arXiv, PubMed, bioRxiv and chemRxiv at once, so a method can lead you to where it is being applied.

AarXiv4 found

PPubMed4 found

BbioRxiv2 found

CchemRxiv2 found

Searching the other archives…