Comparison of the BP training algorithm and LVQ neural networks for e, μ, π identification

Zhang, Z.P; Chen, H.F; Ye, S.W; Zhao, J.

doi:10.1016/0168-9002(96)00733-4

Cited by 5 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, principle component analysis (PCA), linear discriminant analysis (LDA), support vector machine (SVM), and other techniques, are used to reduce the dimensions but maintain the main characters of the data. These algorithms achieve a very high sensitivity and specificity; therefore, they substantiate the potential of Raman spectroscopy to distinguish cancer tissue from normal tissue [15][16][17]. Nevertheless, new algorithms are still needed to lessen the computation time and achieve more accurate diagnosis results.…”

Section: Introductionmentioning

confidence: 77%

Evaluation of Raman spectra of human brain tumor tissue using the learning vector quantization neural network

et al. 2016

View full text Add to dashboard Cite

The Raman spectra of tissue of 20 brain tumor patients was recorded using a confocal microlaser Raman spectroscope with 785 nm excitation in vitro. A total of 133 spectra were investigated. Spectra peaks from normal white matter tissue and tumor tissue were analyzed. Algorithms, such as principal component analysis, linear discriminant analysis, and the support vector machine, are commonly used to analyze spectral data. However, in this study, we employed the learning vector quantization (LVQ) neural network, which is typically used for pattern recognition. By applying the proposed method, a normal diagnosis accuracy of 85.7% and a glioma diagnosis accuracy of 89.5% were achieved. The LVQ neural network is a recent approach to excavating Raman spectra information. Moreover, it is fast and convenient, does not require the spectra peak counterpart, and achieves a relatively high accuracy. It can be used in brain tumor prognostics and in helping to optimize the cutting margins of gliomas.

show abstract

Section: Introductionmentioning

confidence: 77%

Evaluation of Raman spectra of human brain tumor tissue using the learning vector quantization neural network

et al. 2016

View full text Add to dashboard Cite

show abstract

“…A preliminary investigation was conducted to examine the relationship between the recognition window size and the in-control ARL. The in-control ARLs of the proposed CCPR models with recognition window sizes of 16,24,32,40,48, and 56 were 82, 167, 367, 784, 1750, and 4830, respectively. Each in-control ARL value here was obtained based on the average of 10000 independent simulation runs.…”

Section: An Lvq-based Ccpr Model With a Dynamic Training Algorithmmentioning

confidence: 99%

“…Fault diagnosis and classification of signals are two typical applications of an LVQ network. For many tasks, the LVQ network can perform more accurate classifications than the BPN [6,10,35,40]. Moreover, LVQ networks are faster than BPNs during training [6,32].…”

Section: Introductionmentioning

confidence: 98%

Effective Pattern Recognition of Control Charts Using a Dynamically Trained Learning Vector Quantization Network

Guh

Shiue

2008

Journal of the Chinese Institute of Industrial Engineers

View full text Add to dashboard Cite

Unnatural control chart patterns (CCPs) are associated with a particular set of assignable causes for process variation. Hence, effectively recognizing CCPs can substantially narrow down the set of possible causes to be examined, and accelerate the diagnostic search. Recently, machine-learning techniques, especially the artificial neural network (ANN), have been widely used as an effective tool for CCP recognition (CCPR) tasks. Most ANN applications in CCPR have been using static supervised ANNs, such as back propagation networks (BPNs) and learning vector quantization (LVQ) networks. The false recognition problem (i.e. the patterns are misclassified) commonly encountered for these ANN-based CCPR models is mainly due to the fact that the static ANNs cannot appropriately deal with dynamic patterns, such as CCPs. In this research, a dynamic training algorithm is designed to provide an LVQ network-based CCPR model the capability to on-line recognize the dynamic CCPs that vary over time. The numerical results using simulation show that the dynamically trained LVQ network-based model proposed in this research performs much better than other ANN-based models reported in literature with respective to recognition accuracy and speed. Although this research considers the specific application of a real-time CCPR model based on an LVQ network, the proposed dynamic training algorithm could be applied to CCPR systems based on other ANN architectures in general.

show abstract

“…The faster approach is preferred in real-time applications where time is an important factor. In many practical applications, LVQ networks can also perform more accurate classifications than BPNs (Zhang et al 1996, Dieterle et al 2003. Hence, the present work adopted LVQ networks to develop a real-time CCPR system for autocorrelated processes.…”

mentioning

confidence: 99%

Real-time recognition of control chart patterns in autocorrelated processes using a learning vector quantization network-based approach

Guh

2008

International Journal of Production Research

View full text Add to dashboard Cite

Researchers have been investigating the use of artificial neural networks (NNs) in the application of control chart pattern (CCP) recognition with encouraging results in recent years. Most of the NN models in this field are designed to be used in uncorrelated processes where the process data are independent. Unfortunately, the prerequisite of data independence is not even approximately satisfied in many manufacturing processes. To the best of the author's knowledge, no research results have been published to date on the application of NNs for CCP recognition in autocorrelated processes. This work first shows that autocorrelation in process data greatly affects the performance of NN-based CCP recognizers developed with independent data and then presents a learning vector quantization network-based system that can effectively recognize CCPs in real-time for processes with various levels of autocorrelation. The system performance is evaluated in terms of the classification rate and the average run length. An empirical comparison using simulation indicates that the proposed learning-based system performs better than the traditional control chart methods in detecting shifts when the process data are positively correlated, while it also offers pattern classification. A demonstration example is provided using real data.

show abstract

Comparison of the BP training algorithm and LVQ neural networks for e, μ, π identification

Cited by 5 publications

References 5 publications

Evaluation of Raman spectra of human brain tumor tissue using the learning vector quantization neural network

Evaluation of Raman spectra of human brain tumor tissue using the learning vector quantization neural network

Effective Pattern Recognition of Control Charts Using a Dynamically Trained Learning Vector Quantization Network

Real-time recognition of control chart patterns in autocorrelated processes using a learning vector quantization network-based approach

Contact Info

Product

Resources

About