A Progressive Combined Variable Selection Method for Near-Infrared Spectral Analysis Based on Three-Step Hybrid Strategy

Figure 1

Process of variable selection method combined synergy interval partial least squares, iterative variable subset optimization, and bootstrapping soft shrinkage. Several variable groups were firstly collected in rough selection by synergy interval partial least squares. Then some variables were extracted in fine selection by iterative variable subset optimization. Finally, less variables were retained in optimal selection by bootstrapping soft shrinkage.