A modified random forest approach to improve multi-class classification performance of tobacco leaf grades coupled with NIR spectroscopy

Bin, Jun; Ai, Fang-Fang; Fan, Wei; Jiheng, Zhou; Yun, Yong‐Huan; Liang, Yi‐Zeng

doi:10.1039/c5ra25052h

Cited by 41 publications

(15 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SRD algorithm is entirely general and allows ranking, grouping variables, and automatic selection of a set of variables. Liang et al find SRD as “a simple objective procedure.” The background philosophy is indeed quite simple and similar to that for round‐robin tests (proficiency testing): The systematic errors (biases) of different laboratories and/or different measurement techniques follow normal distribution. Naturally, there is no theoretical proof for that, but it is a well‐substantiated empirical finding in analytical chemistry.…”

Section: Methodsmentioning

confidence: 99%

Comparison of validation variants by sum of ranking differences and ANOVA

Héberger

Kollár-Hunek

2019

Journal of Chemometrics

View full text Add to dashboard Cite

The old debate is revived: Definite differences can be observed in suggestions of estimation for prediction performances of models and for validation variants according to the various scientific disciplines. However, the best and/or recommended practice for the same data set cannot be dependent on the field of usage. Fortunately, there is a method comparison algorithm, which can rank and group the validation variants; its combination with variance analysis will reveal whether the differences are significant or merely the play of random errors. Therefore, three case studies have been selected carefully to reveal similarities and differences in validation variants. The case studies illustrate the different significance of these variants well. In special circumstances, any of the influential factors for validation variants can exert significant influence on evaluation by sums of (absolute) ranking differences (SRDs): stratified (contiguous block) or repeated Monte Carlo resampling and how many times the data set is split (5‐7‐10). The optimal validation variant should be determined individually again and again. A random resampling with sevenfold cross‐validations seems to be a good compromise to diminish the bias and variance alike. If the data structure is unknown, a randomization of rows is suggested before SRD analysis. On the other hand, the differences in classifiers, validation schemes, and models proved to be always significant, and even subtle differences can be detected reliably using SRD and analysis of variance (ANOVA).

show abstract

Section: Methodsmentioning

confidence: 99%

Comparison of validation variants by sum of ranking differences and ANOVA

Héberger

Kollár-Hunek

2019

Journal of Chemometrics

View full text Add to dashboard Cite

show abstract

“…Second, given training data, learning the basis vector is called dictionary learning. It can be statistically formulated as (1) In equation 1, b is the training data with samples, A = [a 1 , …, a k ] is defined as a dictionary. a i is one of the dictionary atoms, x and e represent the sparse coefficient vector and the error term respectively, k is the model parameter.…”

Section: Theory Of Src Algorithmmentioning

confidence: 99%

“…The efficiency and the stability are also dissatisfying. 1 Therefore, it is necessary to develop a new method which is fast, high-efficiency and more objective.…”

Section: Introductionmentioning

confidence: 99%

Rapid and Automatic Classification of Tobacco Leaves Using a Hand-Held DLP‑Based NIR Spectroscopy Device

Zhang¹,

Yang

Weijuan³

et al. 2019

J. Braz. Chem. Soc.

View full text Add to dashboard Cite

A hand-held near infrared (NIR) spectroscopy device is much more convenient than a traditional desktop NIR instrument. Thus, it is more suitable for the practical application. An automatic and rapid tool for grading tobacco leaves on the spot using a hand-held digital light processing (DLP)-based NIR spectroscopy device is proposed in this paper. Firstly, the spectral data of the samples is scanned with a hand-held NIR device directly from the tobacco leaves without any samples preparation procedures. Then, the training model of different classes is built and the class of each test sample is predicted by using sparse representation classification (SRC) algorithm. Comparing with the traditional linear discriminant analysis (LDA) and support vector machine (SVM) algorithms, the classification accuracy of SRC method is the highest and has the least computation time. The results show that hand-held NIR spectroscopy technology could be a novel classification tool for grading tobacco leaves in the purchasing on the spot.

show abstract

“…Da et al [18] applied the mixed algorithm of PLS and ANN for the quantitative analysis of the total sugar in tobacco samples. Bin et al [19] proposed a modified random forest approach to improve multi-class classification performance of tobacco leaf grades.…”

Section: Introductionmentioning

confidence: 99%

Improved Deep CNN with Parameter Initialization for Data Analysis of Near-Infrared Spectroscopy Sensors

Wang

Tian

Yang

et al. 2020

Sensors

View full text Add to dashboard Cite

Near-infrared (NIR) spectral sensors can deliver the spectral response of light absorbed by materials. Data analysis technology based on NIR sensors has been a useful tool for quality identification. In this paper, an improved deep convolutional neural network (CNN) with batch normalization and MSRA (Microsoft Research Asia) initialization is proposed to discriminate the tobacco cultivation regions using data collected from NIR sensors. The network structure is created with six convolutional layers and three full connection layers, and the learning rate is controlled by exponential attenuation method. One-dimensional kernel is applied as the convolution kernel to extract features. Meanwhile, the methods of L2 regularization and dropout are used to avoid the overfitting problem, which improve the generalization ability of the network. Experimental results show that the proposed deep network structure can effectively extract the complex characteristics inside the spectrum, which proves that it has excellent recognition performance on tobacco cultivation region discrimination, and it also demonstrates that the deep CNN is more suitable for information mining and analysis of big data.

show abstract

A modified random forest approach to improve multi-class classification performance of tobacco leaf grades coupled with NIR spectroscopy

Abstract: A novel approach, namely MC-UVE-RF, to improve multi-class classification performance of tobacco leaf grades by NIR spectroscopy.

Cited by 41 publications

References 31 publications

Comparison of validation variants by sum of ranking differences and ANOVA

Comparison of validation variants by sum of ranking differences and ANOVA

Rapid and Automatic Classification of Tobacco Leaves Using a Hand-Held DLP‑Based NIR Spectroscopy Device

Improved Deep CNN with Parameter Initialization for Data Analysis of Near-Infrared Spectroscopy Sensors

Contact Info

Product

Resources

About