An improved correlation-based just-in-time modeling method using dynamic partial least squares and adaptive local domain partition

2017 ◽  
Author(s):  
Xiaolong Chen ◽  
Zhizhong Mao ◽  
Runda Jia ◽  
Dong Xiao ◽  
Xiaojun Wang
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
Modeling Method ◽  
Just In Time ◽  
Local Domain ◽  
Time Modeling ◽  
Domain Partition

Characterization of the Thermal Degradation of Vinegar and the Construction of an Identification Model for Chinese Geographical Indication Vinegars by the Py-GC-MS Technique

2017 ◽  
Vol 100 (2) ◽  
pp. 503-509 ◽  
Author(s):  
Cen Xiong ◽  
Zhiyi Su ◽  
Yanjie Zhezng ◽  
Qi Wang ◽  
Yejing Ling ◽  
...  
Keyword(s):  
Data Analysis ◽  
Thermal Degradation ◽  
Least Squares ◽  
Partial Least Squares ◽  
Modeling Method ◽  
Geographical Indication ◽  
Time To Build ◽  
Pls Modeling ◽  
Identification Model

Abstract The pyrolysis (Py)-GC-MS technique was first introduced for the identification of two kinds of Chinese geographical indication vinegars because its advantages are that it is a simple and convenient sample pretreatmentand inlet method. Abundant Py information about vinegars was obtained using Py-GC-MS; 21 common peaks were selected. With the help of the classical partial least-squares (PLS) modeling method for data analysis, two identification models for Shanxi extra-aged (SX) and Zhenjiang (ZJ) vinegars were established, respectively. An N-reducing method was used to select the variables. The variables were reduced one at a time to build the PLS models with the lowest number of misjudgments. Both models had good recognition rates, identifying over 90% of samples correctly. Thus, combining Py-GC-MS and PLS could be regarded as an effective method for the identification of SX and ZJ vinegars.

A gradient descent boosting spectrum modeling method based on back interval partial least squares

2016 ◽  
Vol 171 ◽  
pp. 1038-1046 ◽  
Author(s):  
Dong Ren ◽  
Fangfang Qu ◽  
Ke Lv ◽  
Zhong Zhang ◽  
Honglei Xu ◽  
...  
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
Gradient Descent ◽  
Modeling Method ◽  
Spectrum Modeling
Sign in / Sign up

Export Citation Format

Share Document