Particle Swarm Optimization over Back Propagation Neural Network for Length of Stay Prediction

Suresh, A.; Harish, Kavitha; Radhika, N.

doi:10.1016/j.procs.2015.02.020

Cited by 49 publications

(21 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…38 PSO performs better than BP and GA in terms of rate of concentration. 39,40 According to the above results, we compared our proposed "CLAHE+MGLCM+PSONN" method with the state-of-the-art approaches: NBC, 9 WN+SVM, 10 ELM, 11 and CLAHE+ELM. 12 The results in Table 5 and Figure 8 showed that the "CLAHE+MGLCM+PSONN" method performed the best accuracy of 78.20% in differentiating chronic gingivitis.…”

Section: Discussionmentioning

confidence: 99%

Gingivitis identification via multichannel gray‐level co‐occurrence matrix and particle swarm optimization neural network

Wen

Jiang

Sun

et al. 2019

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The oral maintenance of patients with periodontal disease mainly depends on clinical examination. However, insufficient number of medical workers cannot carry out detailed oral health education for a large number of patients within limited time and provide these patients with proper and effective oral health nursing methods. In this research, our study put forward a new Artificial Intelligence (AI) based method to diagnose chronic gingivitis, which is based on multichannel gray-levelco-occurrence matrix (MGLCM) and particle swarm optimization neural network(PSONN). Meanwhile, different training algorithms were used as comparison groups.The data set contains 800 images: 400 chronic gingivitis images and 400 healthy gingiva images. The results certify that the specificity, sensitivity, precision, accuracy and F1 Score of MGLCM (PSONN as a classifier) method is 78.17%, 78.23%, 78.24% ,78.20%and 78.17%, respectively. The association of MGLCM and PSONN is more accurate and efficient than approaches: NBC, WN+SVM,ELM and CLAHE+ELM. K E Y W O R D Sartificial neural network, gingivitis identification, multichannel gray-level co-occurrence matrix, particle swarm optimization, pattern recognition

show abstract

Section: Discussionmentioning

confidence: 99%

Gingivitis identification via multichannel gray‐level co‐occurrence matrix and particle swarm optimization neural network

Wen

Jiang

Sun

et al. 2019

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…A famous training method is the back-propagation (BP) algorithm. However, the previous paper mentioned the accuracy advantage of the NN trained by PSO when compared with the NN trained by BP algorithm [6][7][8][9]. The previous paper proposed a framework for the co-design of the NN trained by PSO algorithm [15][16][17].…”

Section: Neural Network Trained By Standard Psomentioning

confidence: 99%

“…However, the PSO algorithm has shown the advantages in the training of the NN compared with the using of BP to train the NN. The NN trained by the PSO algorithm had obtained higher accuracy concerning the learning error and the recognition rate than the NN trained by conventional BP algorithm [6][7][8][9]. The PSO is the algorithm based on the social behavior of a swarm.…”

Section: Introductionmentioning

confidence: 99%

Improved PSO Algorithm for Training of Neural Network in Co-design Architecture

Dang¹,

Hoshino²

2019

IJCA

View full text Add to dashboard Cite

This paper proposes a new version of the standard particle swarm optimization (SPSO) algorithm to train a neural network (NN). The improved PSO, called the wPSOd − CV algorithm, is the improved version of the PSOd − CV algorithm presented in a previous study. The wPSOd − CV algorithm is introduced to solve the issue of premature convergence of the SPSO algorithm. The proposed wPSOd − CV algorithm is used in a co-design architecture. Experimental results confirmed the effectiveness of the NN trained by the wPSOd − CV algorithm when compared with the NN trained by the SPSO algorithm and the PSOd − CV algorithm concerning the minimum learning error and the recognition rates.

show abstract

“…BP network is suitable for establishing the nonlinear relationship between the location information of KPCA extraction and the unknown node coordinates, but it is easy to fall into local minimum, slow convergence rate and poor generalization ability. Particle swarm optimization algorithm based on swarm intelligence does not require with the characteristics information of the problem itself, and can effectively shorten the training time of neural network, so using the PSO algorithm to optimize the BP network to find the optimal weights and thresholds of the network before training [23,24]. Particle velocity and position update rule are expressed as follows.…”

Section: Localization Algorithm Based On Pso-bp Neural Networkmentioning

confidence: 99%

A Novel Hybrid Localization Method for Wireless Sensor Network

Wang

Tiansi

et al. 2016

International Journal on Smart Sensing and Intelligent Systems

View full text Add to dashboard Cite

Abstract-Wireless sensor network is a kind of brand-new information acquisition platform, which is realized by the introduction of self-organizing and auto-configuration mechanisms. Node localization technology represents a crucial component of wireless sensor network. In this paper, a localization method based on kernel principal component analysis and particle swarm optimization back propagation algorithm is carefully discussed. First of all, taking KPCA as the front-end system to extract the main components of the localization information, and then regarding the nonlinear principal components extracted from distance vectors as the input samples, and meanwhile taking the coordinates of vertices in addition to the region boundary as the output samples, the PSO-BP neural network is trained to achieve the localization model. Finally the localization of unknown nodes can be estimated. The simulation experiment result showed that the method has high ability of stability and precision, and meets the practical need of localization.Index terms: Wireless sensor network, localization, kernel principal component analysis, particle swarm optimization, back-propagation algorithm.

show abstract

Particle Swarm Optimization over Back Propagation Neural Network for Length of Stay Prediction

Cited by 49 publications

References 10 publications

Gingivitis identification via multichannel gray‐level co‐occurrence matrix and particle swarm optimization neural network

Gingivitis identification via multichannel gray‐level co‐occurrence matrix and particle swarm optimization neural network

Improved PSO Algorithm for Training of Neural Network in Co-design Architecture

A Novel Hybrid Localization Method for Wireless Sensor Network

Contact Info

Product

Resources

About