4. Conclusions
In this study, SPA–LDA displayed better classification performance
compared with SIMCA, particularly in discriminating between
conservation states of caffeinated samples. In fact, the
SPA–LDA model correctly classified all 43 test samples. Moreover,
this model was found to be robust with respect to measurement
noise, as the classification accuracy remained high (96%) even after
the introduction of artificial noise in the spectra. These results suggest
that the proposed methodology is a promising alternative for
assessment of conservation state and decaffeination condition of
coffee samples.
Future works could investigate the possibility of using this
methodology to monitor the ageing process of coffee samples over
time. Moreover, the influence of roasting degree in the classification
results could also be studied.
4. ConclusionsIn this study, SPA–LDA displayed better classification performancecompared with SIMCA, particularly in discriminating betweenconservation states of caffeinated samples. In fact, theSPA–LDA model correctly classified all 43 test samples. Moreover,this model was found to be robust with respect to measurementnoise, as the classification accuracy remained high (96%) even afterthe introduction of artificial noise in the spectra. These results suggestthat the proposed methodology is a promising alternative forassessment of conservation state and decaffeination condition ofcoffee samples.Future works could investigate the possibility of using thismethodology to monitor the ageing process of coffee samples overtime. Moreover, the influence of roasting degree in the classificationresults could also be studied.
การแปล กรุณารอสักครู่..
