|239 Brown Lab||Newark, DE 19716||<div class="ExternalClass026ABD862D404499A7AA47148093FA83"><p style="text-align:justify;"><strong>(b. 1950) B.S., 1972, M.S., 1975, Portland State University; Ph.D., 1978, University of Washington</strong></p></div>||<div class="ExternalClass95E0605F463D494DB46F2CD4AC2B9D0B"><p style="text-align:justify;">Students in my laboratory study the development of new methods for mathematical and statistical analysis of chemical data, a field known variously as chemometrics or chemo-informatics. Most of our current effort is directed to investigating problems in the field of analytical chemistry, but students in this research group are exposed to problems from a wide range of disciplines, ranging from geochemistry to biochemistry and process engineering. Our usual practice is to study real problems provided by collaborators.</p></div>||<div class="ExternalClass6125BB70ABC4430AB107F7836F6EFBFA"><p style="text-align:justify;">We are devising ways to identify and compensate instrumental and background effects in data to improve mathematical methods for relating analytical responses to concentration of analytes. One current application involves developing methods for the transfer of multivariate classifications done on forensic paint samples.</p></div>||<div class="ExternalClass8998465268D244C1A2DF28F3F67FDB38"><p style="text-align:justify;">New methods for extracting information from large amounts of data are increasingly needed as new analytical methods produce more and more data and there is simultaneously a need to focus on finer and finer detail in the instrumental response(s) from a sample, as well as the uncertainty in the multivariate analysis of the data. One current project concerns extraction of geospatial information from data sets made from a combination of isotope ratio and trace metal measurement collected from each sample. A second project aims to fuse information from multiple measurements to extract new information on complex biological samples used in the production of biopharmaceuticals.</p></div>||<div class="ExternalClassCF496EAC1AF0433DBFCAF1969891E9DF"><ul><li><p>D.
Poerio and S.D. Brown, A Frequency-Localized Recursive Partial Least
Squares Ensemble for Soft Sensing, J. Chemom. e2999, 2018. (DOI:
Poerio and S.D. Brown, “Highly-Overlapped, Recursive Partial Least
Squares Soft Sensor with State Partitioning via Local Variable
Selection”,Chemom. Intell. Lab Syst. 175 (2018) 104–115. (DOI: 10.1016/j.chemolab.2018.02.006 )</p></li><li><p>C.
Kneale and S.D. Brown, “Small Moving-Window Calibration Models for Soft
Sensing Processes with Limited History.” Chemom. Intell. Lab. Syst.183, 2018, 36-46. (https://doi.org/10.1016/j.chemolab.2018.10.007)</p></li><li><p>C.
Kneale and S.D. Brown, Band Target Entropy Minimization and Target
Partial Least Squares for Spectral Recovery and Calibration, Analyt.
Chim. Acta, 1031 (2018) 38-46. (DOI:10.1016/j.aca.2018.07.054 )</p></li><li><p>D.
Poerio and S.D. Brown, Localized and Adaptive Soft Sensor Based on an
Extreme Learning Machine with Automated Self-correction Strategies, J.
Chemom., 2018;e3088. (https://doi.org/ 10.1002/cem.3088).</p></li><li><p>C. Kneale and S.D. Brown, Exploratory Data Analysis using an Uncharted Forest, Talanta 189 (2018) 71–78. (DOI: 10.1016/j.talanta.2018.06.061)</p></li></ul></div>||Current Research||Advances in Multivariate Calibration and Classification||Novel Methods for Data Mining, Data Fusion and Knowledge Discovery||Representative Publicationsemail@example.com||Brown, Steven D.||(302) 831-6861||<img alt="" src="/Images%20Bios/brown.jpg" style="BORDER:0px solid;" />||Willis F. Harrington Professor||http://sites.udel.edu/sdb/|