Performance of Classification Analysis: A Comparative Study between PLS-DA and Integrating PCA+LDA

Abstract

Classification methods are fundamental techniques designed to find mathematical models that are able to recognize the membership of each object to its proper class on the basis of a set of measurements. The issue of classifying objects into groups when variables in an experiment are large will cause the misclassification problems. This study explores the approaches for tackling the classification problem of a large number of independent variables using parametric method namely PLS-DA and PCA+LDA. Data are generated using data simulator; Azure Machine Learning (AML) studio through custom R module. The performance analysis of the PLS-DA was conducted and compared with PCA+LDA model using different number of variables (p) and different sample sizes (n). The performance of PLS-DA and PCA+LDA has been evaluated based on minimum misclassification rate. The results demonstrated that PLS-DA performed better than the PCA+LDA for large sample size. PLS-DA can be considered to have a good and reliable technique to be used when dealing with large datasets for classification task.

Multi-element signature of cuttlefish and its potential for the discrimination of different geographical provenances and traceability

“…Therefore, in the context of the present work, S-LDA and LW-LDA, were used for classification by variable selection, but outputs were compared with those obtained through the application of VIP-based PLS-DA. The latter, in fact, consists of a more robust and flexible algorithm, particularly suited for the classification of a large number of samples and is suggested to be a more powerful tool for reliable variable selection compared to the traditional LDA (Rashid, Hussain, Ahmad, & Abdullah, 2019).…”

Sputum Proteome Signatures of Mechanically Ventilated Intensive Care Unit Patients Distinguish Samples with or without Anti-pneumococcal Activity

“…Furthermore, we have performed a linear discriminant analysis (LDA) based on the three groups and compared the obtained results with our PLS analysis. However, the PLS separated the different groups much better than the LDA (data not shown), which probably relates to our high-dimensional MS feature data for which an LDA is not optimal ( 59 – 61 ).…”

Exploring the Steps of Infrared (IR) Spectral Analysis: Pre-Processing, (Classical) Data Modelling, and Deep Learning

“…When combined with supervised classification algorithms, PCA serves as a powerful feature extraction tool [84]. By reducing the dimensionality of the data, PCA enables faster class discrimination based on the actual spectral differences within the dataset instead of relying on pre-assigned spectral differences across classes [85].…”