The use of empirical modelling, namely, feed-forward neural network, is pro
ven to be an efficient alternative for the control of bovine haemoglobin pr
oteolysis from mid-infrared (MIR) spectra recording by the mean of an ATR p
robe. Six batches were experimented and sampled. The situation is challengi
ng since the results of the self-organising maps (SOMs) analysis definitely
show clusters of the spectral dataset. Consequently, strong generalisation
capabilities are to be foreseen and it is typical situation when one deals
with enzymatic processes. Supervised learning is thus used in order to pre
dict an unknown batch from the knowledge provided by the five others. The R
MSEP results are estimated around 0.3% in an analytical range of (0, 8.7) o
f the hydrolysis degree. Moreover, the analysis of the data treatments enab
le to emphasise on the biochemical information, from one hand, and on the r
ole of each of the hidden nodes from another hand. (C) 2001 Elsevier Scienc
e B.V. All rights reserved.