EVRI-thing You Need to Know About Preprocessing

Preprocessing, what you do to data before it goes into a regression or machine learning model, is often the key to modeling success. In this webinar professor Rasmus Bro will demonstrate the great many types of preprocessing that you can do with Eigenvector’s MATLAB based PLS_Toolbox or stand-alone Solo software. You’ll learn how you can combine preprocessing methods in various ways to remove all sorts of artifacts from your data.

Table of Contents

00:00 – Introduction and Notes

04:45 – Reasons to use Preprocessing

07:34 – Preprocessing Example with Mean Centering

12:00 – Preprocessing Example with Autoscale

13:06 – Preprocessing Menu and Show function

19:11 – Preprocessing Example with chromatography data (Autoscale, Pareto, Poisson Scaling, Mean Centering

26:43 – Variable Alignment (COW, Peak Alignment)

29:41 – Preprocessing Example with Arithmetic Operations

34:17 – Preprocessing Example with NIR data and Standard Normal Variate (SNV) and Polynomial and Cross-term transformations

39:08 – iPLS variable selection

44:54 – Orthogonalization Filters and Clutter

46:59 – External Parameter Orthogonalization

48:00 – GLSW – Generalized Least Squares Weighting and OSC

49:30 – Preprocessing Example with Declutter settings

54:44 – Example with Raman Data and Baseline removal

57:10 – Scatter Correction Methods

1:07:00 – Conclusion and Online Resources