Using the Advanced Features of PLS_Toolbox/Solo

April 23, 2013
Cheshire, UK

Eigenvector Research, Inc. is pleased to offer Using the Advanced Features of PLS_Toolbox/Solo at the Advances in Process Analytics and Control Technology (APACT) 2013 conference in Chershire, UK. More information about the course is below.

Target Audience

Using the Advanced Features of PLS_Toolbox/Solo is designed for those with a good understanding of basic chemometrics who wish to explore the power of more advanced chemometric tools. While prior experience with PLS_Toolbox or Solo is not required, the course is especially useful for those who have experience with older versions of Eigenvector software and want to learn about newer features.

Course Description

The course concentrates on "how to" perform advanced chemometric analyses and data manipulations. After an introduction to the basic interface, the course shows how to use PLS_Toolbox and Solo to make and view Principal Components Analysis (PCA) and Partial Least Squares (PLS) regression models. More advanced techniques are then demonstrated including Nonlinear Methods, Advanced Preprocessing, Robust Methods, Multivariate Curve Resolution (MCR) and Calibration Transfer. The course will finish with a question and answer session and time for one-on-one help with the software.

Participants are encouraged to bring their laptops with the latest versions of Solo or PLS_Toolbox installed. Demo copies will work just fine. User with Eigenvector accounts can download free demos. If you don't have an account, start by creating one.

About the Instructors

The course will be led by PLS_Toolbox creator and Eigenvector co-founder Barry M. Wise. Dr. Wise holds a doctorate in chemical engineering and has experience in a wide variety of applications spanning chemical process monitoring, modeling and analytical instrument development. He has extensive teaching experience, having presented over 100 chemometrics courses. He will be assisted by Eigenvector's Vice-President Neal B. Gallagher. Dr. Gallagher has broad experience in a wide variety of chemometric applications spanning analytical instrument and method development, remote sensing, and chemical process monitoring and modeling. He also has extensive teaching experience, including leading courses each spring at Eigenvector University, the most comprehensive offering of chemometrics classes in the world.

Course Fee, Registration and Venue

Course fees, registrations and deadlines are determined by the conference organization. In 2013, the course fees are £150 if you attend the conference and pre-conference courses’ and £300 if you attend this course only’. Please see registration information for APACT 2013.

Schedule

10:00-10:30 Registration and coffee
10:30-12:15 Course
12:15-13:15 Lunch
13:15-14:50 Course
14:50-15:10 Coffee break
15:10-17:00 Course

Course Outline

1. Introduction
1.1 Basic layout of PLS_Toolbox/Solo
1.2 Browse interface
1.3 Analysis interface
1.4 Plot control interface

2. Editing and selecting data
2.1 The DataSet Editor
2.2 Selecting data graphically
2.3 Highlighting samples and variables
2.4 Setting and displaying classes

3. An example with Principal Components Analysis
3.1 Loading data
3.2 Setting preprocessing
3.3 Selecting number of components
3.4 Viewing model parameters: scores, loadings, variance captured

4. An example with Partial Least Squares Regression
4.1 Loading calibration and test data
4.2 Selecting variable ranges
4.3 Setting preprocessing
4.4 Setting cross-validation
4.5 Selecting number of components
4.6 Viewing model parameters
4.7 Options for implementing models on-line

5. Non-linear Methods
5.1 Locally Weighted Regression
5.2 Augmenting data with non-linear terms
5.3 Support Vector Machines

6. Using the Robust Methods
6.1 Robust PCA
6.2 Robust PLS

7. Advanced Preprocessing
7.1 Preprocessing interface
7.2 Selecting and ordering steps, viewing preprocessed data
7.3 Filtering and orthogonalization methods: GLS, EPO, defining clutter
7.4 Looping

8. Orthogonalizing a PLS model--fauxPLS
8.1 Initial model development
8.2 Viewing orthogonalized model

9. Curve Resolution Approaches
9.1 Setting constraints in Multivariate Curve Resolution
9.2 The Purity approach

10. Calibration Transfer Methods
10.1 Piece-wise Direct Standardization
10.2 Generalized Least Squares
10.3 Orthogonal Signal Correction