 |
 |
 |
 |
 |
 |
 |
 |
 |
EigenNews - The Newsletter of Eigenvector Research Vol. 7, No. 1April 5, 2005 EigenNews is an approximately quarterly note about the happenings at Eigenvector Research, Inc., along with news of interest to the chemometrics, analytical chemistry, spectroscopy and process modeling/monitoring communities. (Subscribe and unsubscribe info at bottom.) IN THIS ISSUE: 1) Upcoming Short Courses for 2005
2) PLS_Toolbox News: PLS_Toolbox 3.5.1 Released 3) PLS_Toolbox User Tip: Setting Classes in a DataSet Object 4) NIR Pharmaceutical Table Data Set Available 5) Meeting and Workshop Calendar
6) How did I get on this list? - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Eigenvector will be teaching Process Chemometrics, our popular three-day, hands-on course, in two locations this June. We'll be in Wenatchee, Washington, June 6-8, and at The MathWorks in Natick, MA June 15-17. For more information, please see: http://eigenvector.com/Courses/ProcessChemometrics.html The PRF Summer School on Chemometrics in Sensor Technology will be June 26-July 1 in Atlanta, GA. It is geared towards scientists and engineers active in the development and application of sensor technology. Instructors include: Karl Booksh, Steve Brown, Charles Miller, Boris Mizaikoff, and Neal Gallagher. This course will be held at Georgia Tech Hotel & Conference Center is June 25-July 1. For information write to Boris Mizaikoff at boris.mizaikoff@chemistry.gatech.edu. Rasmus Bro will join us for a series of chemometrics courses in Barcelona, SPAIN, September 12-16, following PLS'05. We'll start with two days of Chemometrics without Equations, and follow with a day each of Variable Selection, Multivariate Image Analysis, and Multivariate Statistical Process Control. For information, please write to bmw@eigenvector.com. We'll also be back at the Federation of Analytical Chemistry and Spectroscopy Societes (FACSS) and Eastern Analytical Symposium (EAS) conferences to teach Chemometrics without Equations and Intro to Multivariate Image Analysis. Please see: http://eigenvector.com/Courses/Conf_CWE.html
or Need chemometrics training at your site? Eigenvector has a rapidly expanding curriculum available. In addition to the courses listed above, we have courses that focus specifically on process monitoring, inferential sensors for use in process control, multivariate image analysis and multi-way analysis. Write to bmw@eigenvector.com for more information. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - PLS_Toolbox 3.5.1 is now available. This minor update to PLS_Toolbox 3.5 is completely compatible with the latest MATLAB release (14 Service Pack 2, MATLAB 7.0.4) and includes a number of improvements and additions. PLS_Toolbox Version 3.5.1 introduces new PURITY analysis methods for self-modeling curve resolution and mixture analysis. Building on the popular SIMPLSMA approach, PURITY is run through the ANALYSIS graphical user interface to perform self-modeling mixture analysis interactively. This allows quick resolution of contributions and spectral components in a data set. The PURITY algorithm also allows the use of second derivatives simultaneously with the conventional spectral data. Also enhanced in PLS_Toolbox Version 3.5.1 are: PCA cross-validation, PLSDA cross-validation and class selection, display options, Excel file import functionality and GRAMS SPC file import compatibility. This release also fixes a variety of version compatibility issues making PLS_Toolbox fully compatible with MATLAB Versions 6.5 through 7.0.4. Compatibility with MATLAB Version 6.1 has also been enhanced. New purchases and upgrades can obtained at: http://software.eigenvector.com/toolbox/3_5/OrderForm.php Want to try PLS_Toolbox 3.5 before you buy? A fully functional demo is available at: http://software.eigenvector.com/toolbox/3_5/OrderForm.php?demo -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 3) PLS_Toolbox User Tip: Setting Classes in a DataSet Object: In many applications, the groups or "classes" of samples in a data set are critical to the modeling and/or interpretation of results. In PCA, scores plots can be quickly interpreted for clustering if samples can be labeled/colored according to class. Likewise, the SIMCA and PLS-DA methods rely on sample class information to generate models and assess results. PLS_Toolbox makes use of the free DataSet object (http://software.eigenvector.com/DataSet) to associate a numerical class to each sample in a given set of data. These classes are automatically used when modeling and viewing your data and models. When viewing scores plots, for example, selecting View/Classes automatically uses unique symbols for each class. Note that in SIMCA and PLS-DA that a class of 0 (zero) represents "unknown class". There are a variety of ways to assign classes to samples using PLS_Toolbox: (1) DataSet Editor: From the Analysis GUI, the DataSet Editor can be opened using the menu item: Edit/X-block Data (similarly, the "editds" command can be used from the MATLAB command line). Because samples are associated with Mode 1 in the data, the classes for samples are assigned in the "Mode 1 Labels" tab. The column headed "Class" contains any classes for the samples. There are four ways of entering classes in the DataSet Editor:
(2) Graphically: When viewing scores or even a plot of your data, you can use the Plot Controls to select points (Click "Select" and drag a box around the points to select), and then choose a numerical class for the selected samples by selecting Edit/Set Class of Selection from the Plot Controls figure. Note that you must be viewing the columns of the data to select classes for samples. (3)
Command-line: Given a DataSet object in the MATLAB workspace,
classes for samples can be set using the command:
where x is the DataSet object and the values between the square brackets [ ] are the list of each sample's class. This information is also available in our FAQ: http://software.eigenvector.com/faq/?44 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 4) NIR Pharmaceutical Tablet Data Set Available In 2002, the International Diffuse Reflectance Conference (IDRC) published a "Shootout" data set consisting of spectra from 654 pharmaceutical tablets from two spectrometers. The data is divided up into calibration, validation and test sets. We've converted it to our MATLAB DataSet Object format for your computing pleasure. It's a great data set for chemometrics training and algorithm testing. Find it at: http://software.eigenvector.com/Data/ -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 5) Meeting and Workshop Calendar The following conferences and workshops should be of interest to chemometricians, analytical chemists and process engineers. If you have a meeting you'd like to list in the next EigenNews, please drop a line to bmw@eigenvector.com
-
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6) How did I get on this list? If you are on this list, it means that you have corresponded with us, gone to one of the same meetings, or have published information about yourself (such as a web page) suggesting that you have an interest in chemometrics, analytical chemistry, spectroscopy, or chemical process modeling and monitoring. In other words, we had some reason to believe that you might actually be interested in receiving this newsletter. Subscribe and Unsubscribe requests can be made on the Eigenvector Subscription Manager page at: http://www.eigenvector.com/subscribe/ If you have any problem getting on or off the list, please write to me directly at bmw@eigenvector.com or contact Eigenvector Research, Inc. directly:
|
|
Eigenvector Research, Inc., 830 Wapato Lake Road, Manson, WA 98831 |