|
About
Eigenvector Research, Inc.
Eigenvector
Research was founded by Barry M. Wise
and Neal B. Gallagher on January
1, 1995. Both co-founders hold doctorates in chemical engineering
and have broad experience modeling chemical systems using both theoretical
and empirical approaches. Jeremy M. Shaver,
Chief of Technology Development, joined Eigenvector in 2001. Shaver
holds a doctorate in analytical chemistry and is especially well
versed in the application of chemometrics to spectroscopy. Chief
Eigenspectroscopist Willem Windig
is especially well known for his work in curve resolution (e.g.
SIMPLISMA, DECRA, CODA). Senior Research Scientist Charles
E. Miller brings extensive process and spectroscopic chemometrics
expertise to Eigenvector. Our most recent addition, Senior Research
Scientist Robert T. Roginski, comes
to us from the pharmaceutical sector where he has been involved
in many aspects of Process Analytical Technology (PAT). Software
Engineer R. Scott Koch, specialist
in software management and database design, rounds out the technical
staff. Our experienced staff makes Eigenvector Research well prepared
to apply modern multivariate statistical methods to your data while
keeping a firm grasp of the underlying chemical, physical and biological
aspects of the system.
Want
a layman's explanation of our work? See what our local paper, the
Lake Chelan Mirror,
had to say about us in their June 13, 2001 article, "They've
Got Chemistry."
Software
Products
Eigenvector
Research is owned and operated by the authors of the PLS_Toolbox
for use with MATLABŪ. The PLS_Toolbox is the most advanced chemometrics
package available today. Our experience in developing end-user software
translates into powerful tools for your special applications.
Services
We offer a wide range of chemical data analysis and custom application
development consulting services. This includes applications of both
established and state-of-the-art chemometric techniques. Eigenvector
Research can also work with you to develop new analytical instruments
and processes and to develop custom models and software for process
monitoring, fault and upset detection, and dynamic models for process
control purposes.
Data Modeling Techniques
Eigenvector Research has experience with a wide variety of process
data modeling techniques including:
- Data
exploration: Principal Components Analysis (PCA), and cluster
analysis
- Linear
regression: Partial Least Squares (PLS), Principal Components
Regression (PCR) and Continuum Regression
- Non-linear
modeling: Neural Net-PLS and Locally Weighted Regression
- Genetic
Algorithms: GAs for variable and non-linear model structure selection
- Second-order
calibration: Generalized Rank Annihilation, Multivariate Curve
Resolution and Evolving Factor Analysis
|