Host Dr. Jerry Workman speaks with Dr. Brian G. Rohrback to discuss to his research and experience in automating the process of building multivariate calibrations.
Enjoy the podcast. For more information and further conversation, contact us at info@infometrix.com.
Click to access Analytically Speaking Podcast Ep.33
Expo Manufactura 2025
February 11-13, 2025
Av. Constitucion Oriente : 300, Col. Centro, Monterrey, NLE, 64000, Mexico
Connect and discover the latest in manufacturing. Expo Manufactura is just around the corner. Come visit Infometrix, exhibiting with Washington State Department of Commerce in Stand #830. Hope to see you there.
For more info, contact info@infometrix.com.
The Analyzer Technology Conference for 2025 is coming up in April and is likely the best source of information for process instrumentation and application reviews in the chemical and petroleum industries. Infometrix has been describing the automation of chemometric calibrations in this forum since 2015 independently and in concert with Chevron and Yokogawa. Most recently, the fundamentals of quality control employing optical spectroscopy were presented and the paper is available for download.
Click to access rohrback-atc-2024-paper.pdf
Chemometric plus Automation equals Machine Learning and Best Practices.
A calibration model tends to improve as additional calibration samples are added to the library. If the samples reflect variation in the chemical composition, the model then expands its zone of relevance. Ultimately, as these new zones expand calibration scope, the model can degrade in performance due to nonlinearities and may require adding to the model rank, which can make the model fragile. Building local models is one answer to gain the best of both worlds, but optimizing a locally weighted model is tricky and time consuming. Automating the optimization of locally weighted models is a solution.
http://doi.org/10.1002/cem.3637
For additional information or questions, contact info@infometrix.com.
#chemometrics #LWR #spectroscopy #regression
Venue: Raleigh Convention Center
Date: Oct. 25, 2024
Time: 3:50pm – 4:10pm EST USA Time
Location: 302C
Presented by: Brian Rohrback, President, Infometrix, Inc.
Abstract: Chemometrics has played a critical role in the supervised learning realm by supplying multivariate tools to extract information content from streams of data. Machine learning is not new, enjoying origins in the 1950s as applied to game theory. At its core, the purpose is to replace a human-centered process with software that can adapt to new inputs in an optimal, sometimes novel, way. Adapting chemometrics to fit the machine learning paradigm necessitates applying the tools of the trade in a fully automated way and, in quality assurance applications, the focus needs to be on spectroscopy (optical, magnetic, mass) and chromatography; these are the most direct measures of chemistry in the system under management. This paper shines light on the process of integrating machine learning and chemometrics for analytical instruments.
Register at SciX 2024.
infometrix.com 