Efficient wavenumber selection based on nearest correlation Louvain method for NIR calibration modeling

In process analytical technology (PAT), partial least squares (PLS) regression has been widely used to construct calibration models for near-infrared (NIR) spectroscopy. To construct a highly accurate calibration model, wavenumber selection is crucial. In the present work, an efficient wavenumber selection method especially for PLS is proposed. The proposed method is referred to as nearest correlation Louvain method-based variable selection (NCLM-VS). NCLM-VS is a correlation-based group-wise method; it constructs an affinity matrix of input variables by the nearest correlation (NC) method, partitions the affinity matrix by the Louvain method (LM), consequently clusters input variables into multiple variable groups, and finally selects variable groups according to their contribution to estimates. LM is very fast and optimizes the number of groups automatically unlike spectral clustering (SC). The advantage of NCLM-VS over conventional methods including nearest correlation spectral clustering-based method (NCSC-VS) is demonstrated through their applications to calibration modeling based on near-infrared (NIR) spectra. In particular, it is confirmed that NCLM-VS is significantly faster than NCSC-VS while NCLM-VS can achieve as good estimation performance as NCSC-VS.