Support vector regression for determination of component of compound oxytetracycline powder on near-infrared spectroscopy

Zou, Tingting; Dou, Yong; Mi, Hong; Zou, Jiying; Ren, Yuping

doi:10.1016/j.ab.2006.04.025

Cited by 42 publications

(27 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ANN has been shown to give better results than PLS and PCR in regression analysis and it is particularly regarded for its ability to model non-linear relationships in the data [12]. However, chemometric techniques have not always been found to be inferior to the ANN [13]. A key problem with the ANN is that their use is considered by many to be a "black art": finding the right network structure (number of hidden nodes, type of threshold function) and the selection of initial connection weights can be a problem in the generation of an ANN model, all of which has a direct impact on the performance achieved.…”

Section: Reviewmentioning

confidence: 99%

“…However, some researchers have applied machine learning methods to this domain, for example: decision trees [10]; Naïve Bayes [11]; Artificial Neural Network (ANN) [12]; and Support Vector Machine (SVM) [13]. Some machine learning methods appear to be unsuited to dealing with spectral data, including Naïve Bayes, because of its independence assumption, and k-NN, which does not work as well in high-dimensional spaces.…”

Section: Reviewmentioning

confidence: 99%

See 1 more Smart Citation

A Machine Learning Application for Classification of Chemical Spectra

Madden

Howley²

2009

Applications and Innovations in Intelligent Systems XVI

View full text Add to dashboard Cite

Abstract. This paper presents a software package that allows chemists to analyze spectroscopy data using innovative machine learning (ML) techniques. The package, designed for use in conjunction with lab-based spectroscopic instruments, includes features to encourage its adoption by analytical chemists, such as having an intuitive graphical user interface with a step-by-step 'wizard' for building new ML models, supporting standard file types and data preprocessing, and incorporating well-known standard chemometric analysis techniques as well as new ML techniques for analysis of spectra, so that users can compare their performance. The ML techniques that were developed for this application have been designed based on considerations of the defining characteristics of this problem domain, and combine high accuracy with visualization, so that users are provided with some insight into the basis for classification decisions.

show abstract

Section: Reviewmentioning

confidence: 99%

Section: Reviewmentioning

confidence: 99%

A Machine Learning Application for Classification of Chemical Spectra

Madden

Howley²

2009

Applications and Innovations in Intelligent Systems XVI

View full text Add to dashboard Cite

show abstract

“…The experimental results on synthetic data and real financial data demonstrate its advantages over the standard SVR [22]. In addition, Zou et al compared three regression approaches, including SVR, Artificial Neural Networks (ANNs), and Partial Least Squares (PLS), in quantitative analysis of components of solid pharmaceutical samples on near-infrared spectroscopy, and the results showed that SVR obtained better performance in terms of parameter-selecting process than those obtained with the traditional methods [23]. Although utilizing SVR in time series prediction has yielded many successful results, there has been far less research on the application of SVR to demand forecasting problems.…”

Section: Introductionmentioning

confidence: 99%

A Hybrid GA-TS Algorithm for Optimizing Networked Manufacturing Resources Configuration

Zhan¹,

Lu²,

Li³

2013

Appl. Math. Inf. Sci.

View full text Add to dashboard Cite

Abstract:In recent years, more and more researchers have focused on various significant issues of resource configuration, such as maximum computing performance and minimum response time, especially taking existing resource utilization into account. The goal of this research is to design a resource configuration optimization system under networked manufacturing environment. A prediction mechanism is realized by using support vector regression (SVR) to estimate resource utilization according to network protocol of each manufacturing process, while redistributing resources based on the current status of existing resources pool. A resource configuration mechanism applying genetic algorithm working along with the tabu search process (GA-TS) is proposed in this study to determine the redistribution of resources. The experimental results show that the proposed scheme achieves an effective configuration via reaching the balance between the utilization of resources within existing resources and network protocol of each manufacturing process between potential resource requirements and the network resource providers.

show abstract

“…There exist a number of excellent introductions of SVM. [23][24][25][26][27][28][29][30][31][32][33][34] The theory of LS-SVM has also been described clearly by Suykens et al 21,22 Applications of LS-SVM in quantification and classification have been reported by some workers. [35][36][37][38][39] So, we will only briefly describe the theory of LS-SVM.…”

mentioning

confidence: 99%

Spectrophotometric Determination of Ternary Mixtures of Thiamin, Riboflavin and Pyridoxal in Pharmaceutical and Human Plasma by Least-Squares Support Vector Machines

2007

View full text Add to dashboard Cite

Vitamins are essential compounds in living systems; they differ in their chemical structure and physiological action. Analytical methods have been developed for vitamin identification and/or quantification, using a wide variety of strategies. Thus, the food and pharmaceutical industries have taken advantages of these reliable methods and used them for the estimation of vitamins from simple to complex matrices. Mixtures of vitamins such as vitamin B complex 1,2 and multivitamins in tables and other pharmaceutical formulations are used in the treatment of several diseases.Therefore, the simultaneous determination of mixtures of vitamins is very useful in the pharmaceutical industry. Several methodologies have been developed for the determination of vitamins in different samples, such as liquid chromatography, 3-5 high performance liquid chromatography, 6 micellar electrokinetic capillary chromatography 7 and spectrophotometry and derivative methods. 8,9 The greatest difficulties with simultaneous spectrophotometric determination methods arise when the analytes to be determined give partly or fully overlapped spectra, as is the case with the ingredients of most pharmaceutical preparations. Nowadays, the combination of chemometrics methods with the computer controlled instruments to monitor the molecular absorption spectra creates powerful methods in multicomponent analysis. 10 Partial least squares regression (PLSR) is the most commonly used multivariate calibration method. It is based on linear models and is used as a satisfactory solution in most cases where a linear relationship is present between the spectra and the property to be determined (concentration, for example). 10 However, PLSR is not always the best option, especially in situations where a nonlinear model is clearly required. Theory and application of PLS have been discussed by several workers. [10][11][12][13][14][15][16][17][18] A support vector machine (SVM) is an algorithm from the machine learning community, developed by Cortes and coworker.19 Due to its remarkable generalization performance, SVM has attracted attention and gained extensive application in pattern recognition and regression problems. 20 SVM maps input data into a high dimensional feature space where it may become linearly separable by a hyperplane. One reason that SVM often performs betters than other methods is that SVM was designed to minimize structural risk; such a design has been shown to be superior to the traditional empirical risk minimization principle employed by conventional neural networks.Especially, Suykens and coworker 21,22 proposed a modified version of SVM called least-squares SVM (LS-SVM), which resulted in a set of linear equations instead of a quadratic programming problem, which can extend the applications of the SVM. There exist a number of excellent introductions of SVM. [23][24][25][26][27][28][29][30][31][32][33][34] The theory of LS-SVM has also been described clearly by Suykens et al. 21,22 Applications of LS-SVM in quantification and classification ha...

show abstract

Support vector regression for determination of component of compound oxytetracycline powder on near-infrared spectroscopy

Cited by 42 publications

References 24 publications

A Machine Learning Application for Classification of Chemical Spectra

A Machine Learning Application for Classification of Chemical Spectra

A Hybrid GA-TS Algorithm for Optimizing Networked Manufacturing Resources Configuration

Spectrophotometric Determination of Ternary Mixtures of Thiamin, Riboflavin and Pyridoxal in Pharmaceutical and Human Plasma by Least-Squares Support Vector Machines

Contact Info

Product

Resources

About