What's The View from Eigenvector? (It's our blog!) Follow Us on Twitter

Eigenvector Research, Inc. (EVRI) is a Chemometrics Research and Applications Company located in Wenatchee, Washington, USA. Our mission is to provide advanced chemometrics support for the semiconductor and chemical process industries, consumer product manufacturers and analytical instrument developers. Our goal is to be your complete source for state-of-the-art chemometric tools and know-how.

Below you will find the latest news on our courses, software, and consulting offerings. Use the information on the right to contact us for more information.

Quick Contents: PLS_Toolbox 5.0 / Short Course News / New Tools for On-line Predictions / Other Stuff

PLS_Toolbox 5.2 and our stand-alone product Solo 5.2 were released on April 7. We've added some new functionality plus a host of improvements "under the hood" that make PLS_Toolbox/Solo more efficient and flexible.

Our main model-building interface, Analysis (shown at left), has been updated to allow much more flexible data loading and management. You can now view how your model applies to test data instantaneously. New visualization tools make it easier to find trends in your data and quickly identify what those trends mean. Plus the "linking" between data, variables and samples has been improved to help you sort through your data.

Many of our standard features have been enhanced including an improved Model Cache with additional data management features and new Model Robustness Testing tools (shown below right) which allow you to determine how your model will hold up over time.

Other new features in version 5.0 included the addition of new data importing options (including improved text-file importing and support for Hamilton Sundstrand ASF, AIF, and BKH files), new preprocessing methods and visualization, new clustering method, dbscan, and an automatic creation of model building scripts to reproduce any model.

Not familiar with Solo? Our stand-alone product, Solo, lets you use the main analysis tools from PLS_Toolbox in a purely point-and-click environment. Solo is powered by MATLAB® libraries, so you get MATLAB's numerical accuracy and speed. But you don't have to own MATLAB to use it. That's why we call it Solo! And all the new point-and-click features we add to PLS_Toolbox go directly into Solo as well.

Solo lets you do PCA, Multi-way PCA, CLS, PLS regression, PLS discriminant analysis, and even curve resolution and PARAFAC in the same intuitive interface. Like PLS_Toolbox, Solo allows users to edit data sets either graphically or in a familiar spreadsheet-like interface. Data can be imported from a variety of sources, including spreadsheets, MATLAB .mat files, Galactic .spc files, and delimited ASCII files. Solo uses the same DataSet Object format as Eigenvector's other MATLAB toolboxes.

Current users get PLS_Toolbox 5.2 or Solo 5.2 now from your account.

Short Course News

Thanks to all the participants at the 2009 edition of Eigenvector University! We hope you enjoyed the week as much as we did. See The View from Eigenvector for more news from EigenU 2009. Our next EigenU is tentatively scheduled for May 16-21, 2010.

EigenU Redux! We're doing a Basic Chemometrics course and a day of Special Topics here in Wenatchee, WA, July 13-16. See the blog post for details.

Chemometrics in SIMS will be presented in Toronto, Canada, on September 15, 2009. The course is being held in conjunction with the SIMS XVII conference.

Eigenvector will present Chemometrics without Equations and Advanced Chemometrics without Equations at FACSS 2009, October 18-22, in Louisville, KY.

We'll be back at EAS (Somerset, NJ, November 16-19, 2009) with our one day version of Chemometrics without Equations. Eigenvector Associate Don Dahlberg will lead the course on Sunday, November 15 (date tentative). Registration will be handled by EAS.

Have a short course topic you'd like us to consider? Write to bmw@eigenvector.com.

New Products for On-Line Predictions:
Solo_Predictor and Model_Exporter

We're pleased to announce the release of two new products for putting chemometric models on-line, Solo_Predictor and Model_Exporter. Both of these tools can be used to put models developed in PLS_Toolbox and Solo on-line, but in different ways. As its name implies, Solo_Predictor is a stand alone prediction engine that can be used directly with PLS_Toolbox or Solo models. Model_Exporter, on the other hand, translates PLS_Toolbox and Solo models into numerical recipes that can be imbedded into your own applications. The features of each are listed below.

Stand alone prediction engine for PLS_Toolbox/Solo models Supports PCA, PLS, PCR, CLS, PLS-DA, PARAFAC, etc... Handles missing data, variable alignment Supports advanced preprocessing methods TCP/IP socket and ActiveX interfaces Supports calibration transfer/instrument standardization For more info, please see the Solo_Predictor page

Converts PLS_Toolbox/Solo models into simple numerical recipe Output is easily parsable text script Works with PCA/PCR/PLS models with preprocessing No external libraries required Output formats include: - XML - Matlab m-file (for use in MATLAB or Octave) - Tcl script (for use in Tcl or Symbion) For more info, please see the Model_Exporter page

Other Stuff...

Search Eigenvector Site:

The View from Eigenvector: EVRI's new blog, where we post news of interest, comments on developments in chemometrics, news for our users, op-ed pieces, etc. Stop by and see how things look from here! For instance, see the History of PLS_Toolbox.

EigenNews: Check out back issues of our newsletter EigenNews: April 5, 2005, August 31, 2004, January 22, 2004, September 12, 2003, April 7, 2003, December 12, 2002, January 9, 2002, March 11, 2001, December 15, 2000, August 8, 2000, May 27, 2000, January 6, 2000, October 18, 1999 and July 7, 1999. Note that EigenNews has been replaced by our blog, The View from Eigenvector.

White Papers : Curious about ways people are using Eigenvector Research software products? Check out our white papers. These briefs give some examples of how folks are putting our tools to work for them.

Discussion Lists: Eigenvector supports several discussion and announcement lists. To subscribe (or unsubscribe) to any of our lists, visit our List Subscription Manager.

DataSet Object: Eigenvector manages the free open-source DataSet Standard Data Object for use with MATLAB. The DataSet Object makes it easier to handle and transfer complete data sets in MATLAB.

Data Archive: Need some data to test out your latest algorithm? Check out the Eigenvector Data Archive.

Directions to Eigenvector: Need to find us, or just want to know how to get to Lake Chelan? Directions for Getting to Manson are available, including Dining Suggestions for the trip.

Past Conference Info: We still have information available about CAC 2002, TRICAP'97 (including the group photo and the pictures from our trip to Stehekin), and the Chemometrics Sessions at FACSS'98.

Review our Privacy Policy.

There is always something going on at EVRI. To keep up with the latest developments, or to be included in special events and promotions, please drop us a line at bmw@eigenvector.com.