Prediction of the Freshness of Grass Carp during Storage with Electric Nose Based on Signal Sequence Merging and Wavelet Transform
Table 1
Prediction results of PLS and ANN models based on signal sequence merging and WT for TVB-N content in grass carp fillet.
DSSM
RSSM
ANN
PLS
ANN
PLS
RMSECV (mg/100 g)
RMSECV (mg/100 g)
RMSECV (mg/100 g)
RMSECV (mg/100 g)
0.9526
2.0350
0.9465
2.1411
0.8984
2.9174
0.9465
2.1411
0.9566
1.9592
0.9454
2.1638
0.9614
1.8627
0.9440
2.1899
0.9446
2.2275
0.9435
2.1998
0.9567
1.9608
0.9427
2.2149
0.9683
1.6582
0.9406
2.2541
0.9561
1.9777
0.9392
2.279
0.9404
2.3046
0.9296
2.4463
0.9574
1.9326
0.9343
2.3657
0.9474
2.1423
0.9257
2.5097
0.9598
1.8835
0.9227
2.5586
0.9459
2.1989
0.9454
2.1638
0.9445
2.1124
0.9440
2.1899
0.8727
3.5193
0.9685
1.6521
0.7886
4.7766
0.9432
2.2051
0.8641
3.6728
0.9538
1.9937
0.8663
3.5316
0.9420
2.2269
0.9304
2.4694
0.9606
1.8437
0.9171
2.7992
0.9505
2.0628
0.9127
2.7949
0.9275
2.4801
0.9151
2.7579
0.8839
3.1035
The four models demonstrate outstanding performance in TVB-N content predicition and were subsequently discussed collectively. R2 and RMSECV are highlighted in bold to facilitate quick identification of the models by other researchers.