A classification tool for N-way arrays based on SIMCA methodology
A classification tool for N-way arrays based on SIMCA methodology
A classification tool for N-way arrays based on SIMCA methodology
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Final remarks<br />
Perspective<br />
‣ sampling is a more critical issue than limits criteria<br />
‣ choice of dimensi<strong>on</strong>ality as low RMSECV suboptimal with respect to<br />
SENS/SPEC compromise<br />
‣ empirical CV 95% promising c<strong>on</strong>sidering it is simple<br />
‣ there may be cases where fit per<str<strong>on</strong>g>for</str<strong>on</strong>g>ms better then CV and viceversa<br />
independently of the <str<strong>on</strong>g>way</str<strong>on</strong>g> limits are estimated<br />
‣ test <strong>on</strong> more data sets and applicati<strong>on</strong>s needed<br />
‣ code needs some refinement, availability at www.life.model.dk<br />
‣ implement a parameter analogous to the variable discriminant power,<br />
<str<strong>on</strong>g>for</str<strong>on</strong>g> model interpretati<strong>on</strong><br />
‣ test use of <strong>on</strong>ly a reference distributi<strong>on</strong> <str<strong>on</strong>g>for</str<strong>on</strong>g> combined Q and D<br />
‣ robust decompositi<strong>on</strong> methods<br />
TRICAP 2009<br />
Dr. Marina CocchiMORE chemometrics,<br />
Department of Chemistry-University of Modena