Pattern Recognition of Vis/NIR Spectroscopy from White Vinegar Based on PLS and BP-ANN Model

Due to more and more tea varieties in the current tea market, rapid and accurate identification of tea (<I>Camellia sinensis</I> L.) varieties is crucial to the tea quality control. Fourier Transform Near-Infrared (FT-NIR) spectroscopy coupled with the pattern recognition was used to identify individual tea varieties as a rapid and non-invasive analytical tool in this work. Seven varieties of Chinese tea were studied in the experiment. Linear Discriminant Analysis (LDA) and Artificial Neural Network (ANN) were compared to construct the identification models based on Principal Component Analysis (PCA). The number of principal components factors (PCs) was optimised in the constructing model. The experimental results showed that the performance of ANN model was better than LDA models. The optimal ANN model was achieved when four PCs were used, identification rates being all 100% in the training and prediction sets. The overall results demonstrated that FT-NIR spectroscopy technology with ANN pattern recognition method can be successfully applied as a rapid method to identify tea varieties.

Download Full-text

Artificial Neural Networks and Their Applications in Business

Encyclopedia of Information Science and Technology, Fourth Edition ◽

10.4018/978-1-5225-2255-3.ch576 ◽

2018 ◽

pp. 6642-6657

Author(s):

Trevor J. Bihl ◽

William A. Young II ◽

Gary R. Weckman

Keyword(s):

Neural Networks ◽

Pattern Recognition ◽

Big Data ◽

Artificial Neural Networks ◽

Model Complexity ◽

Ann Model ◽

Processing Methods ◽

Starting Point ◽

Clustering And Classification ◽

Artificial Neural

Despite the natural advantage humans have for recognizing and interpreting patterns, large and complex datasets, as in Big Data, preclude efficient human analysis. Artificial neural networks (ANNs) provide a family of pattern recognition approaches for prediction, clustering and classification applicable to KDD with ANN model complexity ranging from simple (for small problems) highly complex (for large issues). To provide a starting point for readers, this chapter first describes foundational concepts that relate to ANNs. A listing of commonly used ANN methods, heuristics, and criteria for initializing ANNs is then discussed. Common pre- and post- data processing methods for dimensionality reduction and data quality issues are then described. The authors then provide a tutorial example of ANN analysis. Finally, the authors list and describe applications of ANNs to specific business related endeavors for further reading.

Download Full-text

Artificial Neural Networks and Their Applications in Business

Advanced Methodologies and Technologies in Artificial Intelligence, Computer Simulation, and Human-Computer Interaction - Advances in Computer and Electrical Engineering ◽

10.4018/978-1-5225-7368-5.ch074 ◽

2019 ◽

pp. 1009-1027

Author(s):

Trevor J. Bihl ◽

William A. Young II ◽

Gary R. Weckman

Keyword(s):

Neural Networks ◽

Pattern Recognition ◽

Big Data ◽

Artificial Neural Networks ◽

Model Complexity ◽

Ann Model ◽

Processing Methods ◽

Starting Point ◽

Clustering And Classification ◽

Artificial Neural

Despite the natural advantage humans have for recognizing and interpreting patterns, large and complex datasets, as in big data, preclude efficient human analysis. Artificial neural networks (ANNs) provide a family of pattern recognition approaches for prediction, clustering, and classification applicable to KDD with ANN model complexity ranging from simple (for small problems) to highly complex (for large issues). To provide a starting point for readers, this chapter first describes foundational concepts that relate to ANNs. A listing of commonly used ANN methods, heuristics, and criteria for initializing ANNs are then discussed. Common pre- and post-data processing methods for dimensionality reduction and data quality issues are then described. The authors then provide a tutorial example of ANN analysis. Finally, the authors list and describe applications of ANNs to specific business-related endeavors for further reading.

Download Full-text

Prediction of Sensory Parameters of Cured Ham: A Study of the Viability of the Use of NIR Spectroscopy and Artificial Neural Networks

Sensors ◽

10.3390/s20195624 ◽

2020 ◽

Vol 20 (19) ◽

pp. 5624

Author(s):

Pedro Hernández-Ramos ◽

Ana María Vivar-Quintana ◽

Isabel Revilla ◽

María Inmaculada González-Martín ◽

Miriam Hernández-Jiménez ◽

...

Keyword(s):

Neural Network ◽

Nir Spectroscopy ◽

Sensory Profile ◽

Training Algorithm ◽

Ann Model ◽

Fiber Optic Probe ◽

Organoleptic Characteristics ◽

Dry Cured Ham ◽

Artificial Neural ◽

Sensory Parameters

Dry-cured ham is a high-quality product owing to its organoleptic characteristics. Sensory analysis is an essential part of assessing its quality. However, sensory assessment is a laborious process which implies the availability of a trained tasting panel. The aim of this study was the prediction of dry-ham sensory characteristics by means of an instrumental technique. To do so, an artificial neural network (ANN) model for the prediction of sensory parameters of dry-cured hams based on NIR spectral information was developed and optimized. The NIR spectra were obtained with a fiber-optic probe applied directly to the ham sample. In order to achieve this objective, the neural network was designed using 28 sensory parameters analyzed by a trained panel for sensory profile analysis as output data. A total of 91 samples of dry-cured ham matured for 24 months were analyzed. The hams corresponded to two different breeds (Iberian and Iberian x Duroc) and two different feeding systems (feeding outdoors with acorns or feeding with concentrates). The training algorithm and ANN architecture (the number of neurons in the hidden layer) used for the training were optimized. The parameters of ANN architecture analyzed have been shown to have an effect on the prediction capacity of the network. The Levenberg–Marquardt training algorithm has been shown to be the most suitable for the application of an ANN to sensory parameters

Download Full-text

Classification of Chinese Soybean Paste by Fourier Transform Near-Infrared (FT-NIR) Spectroscopy and Different Supervised Pattern Recognition

Food Analytical Methods ◽

10.1007/s12161-011-9331-0 ◽

2011 ◽

Vol 5 (4) ◽

pp. 928-934 ◽

Cited By ~ 15

Author(s):

Hui Jiang ◽

Guohai Liu ◽

Xiahong Xiao ◽

Shuang Yu ◽

Congli Mei ◽

...

Keyword(s):

Pattern Recognition ◽

Fourier Transform ◽

Near Infrared ◽

Nir Spectroscopy ◽

Soybean Paste

Download Full-text

Detection of the BRAF V600E Mutation in Colorectal Cancer by NIR Spectroscopy in Conjunction with Counter Propagation Artificial Neural Network

Molecules ◽

10.3390/molecules24122238 ◽

2019 ◽

Vol 24 (12) ◽

pp. 2238 ◽

Cited By ~ 1

Author(s):

Xue Zhang ◽

Yang Yang ◽

Yalan Wang ◽

Qi Fan

Keyword(s):

Neural Network ◽

Colorectal Cancer ◽

Artificial Neural Network ◽

Nir Spectroscopy ◽

Braf V600e Mutation ◽

Braf V600e ◽

Ann Model ◽

Wild Type ◽

Tissue Samples ◽

Cp Ann

This paper proposes a sensitive, sample preparation-free, rapid, and low-cost method for the detection of the B-rapidly accelerated fibrosarcoma (BRAF) gene mutation involving a substitution of valine to glutamic acid at codon 600 (V600E) in colorectal cancer (CRC) by near-infrared (NIR) spectroscopy in conjunction with counter propagation artificial neural network (CP-ANN). The NIR spectral data from 104 paraffin-embedded CRC tissue samples consisting of an equal number of the BRAF V600E mutant and wild-type ones calibrated and validated the CP-ANN model. As a result, the CP-ANN model had the classification accuracy of calibration (CAC) 98.0%, cross-validation (CACV) 95.0% and validation (CAV) 94.4%. When used to detect the BRAF V600E mutation in CRC, the model showed a diagnostic sensitivity of 100.0%, a diagnostic specificity of 87.5%, and a diagnostic accuracy of 93.8%. Moreover, this method was proven to distinguish the BRAF V600E mutant from the wild type based on intrinsic differences by using a total of 312 CRC tissue samples paraffin-embedded, deparaffinized, and stained. The novel method can be used for the auxiliary diagnosis of the BRAF V600E mutation in CRC. This work can expand the application of NIR spectroscopy in the auxiliary diagnosis of gene mutation in human cancer.

Download Full-text

Rapid Analysis of Geographical Origins and Age ofTorreya grandisSeeds by NIR Spectroscopy and Pattern Recognition Methods

Journal of Spectroscopy ◽

10.1155/2013/468509 ◽

2013 ◽

Vol 2013 ◽

pp. 1-9

Author(s):

Lu Xu ◽

Si-Min Yan ◽

Chen-Bo Cai ◽

Wei Zhong ◽

Xiao-Ping Yu

Keyword(s):

Pattern Recognition ◽

Sensitivity And Specificity ◽

Near Infrared ◽

Nir Spectroscopy ◽

Classification Performance ◽

Model Complexity ◽

Pattern Recognition Methods ◽

Torreya Grandis ◽

Second Derivative Spectra ◽

Derivatives Of

This paper develops a rapid method for discriminating the geographical origins and age of roastedTorreya grandisseeds by near infrared (NIR) spectroscopic analysis and pattern recognition. 337 samples were collected from three main producing areas and produced in the last two years. The objective of geographical origins analysis is to discriminate the seeds from Fengqiao with a protected geographical indication (PGI) from those of another two provinces. Age classification is aimed to detect the old seeds produced in the last year from the freshly produced ones. Partial least squares discriminant analysis (PLSDA) was used to develop classification models, and the influence of data preprocessing methods on classification performance was also investigated. Taking second-order derivatives of the raw spectra proves to be the most proper and effective preprocessing method, which can remove baselines and backgrounds and reduce model complexity. With second derivative spectra, the sensitivity and specificity were 0.939 and 0.871 for age discrimination, respectively. Perfect classification was obtained, and both sensitivity and specificity were 1 for discrimination of geographical origins.

Download Full-text