Interpreting and Reporting Principal Component Analysis in Food Science Analysis and Beyond
- 149 Downloads
Principal component analysis (PCA) is one of the most widely used data mining techniques in sciences and applied to a wide type of datasets (e.g. sensory, instrumental methods, chemical data). However, several questions and doubts on how to interpret and report the results are still asked every day from students and researchers. This brief communication is inspired in relation to those questions asked by colleagues and students. Please note that this article is a focus on the practical aspects, use and interpretation of the PCA to analyse multiple or varied data sets. In summary, the application of the PCA provides with two main elements, namely the scores and loadings. The scores provide with a location of the sample where the loadings indicate which variables are the most important to explain the trends in the grouping of samples.
KeywordsPrincipal components Scores Loadings Data sets
The authors thank the support of our colleagues and friends that encouraged writing this article.
Compliance with Ethical Standards
Conflict of Interest
Dr. Daniel Cozzolino declares that he has no conflict of interest. Dr. Aoife Power declares that she has no conflict of interest. Dr. James Chapman declares that he has no conflict of interest.
This article does not contain any studies with human or animal subjects.
(In case humans are involved) Informed consent was obtained from all individual participants included in the study. (If not applicable on the study) Not applicable.
- Brereton RG (2008) Applied chemometrics for scientist. Wiley, ChichesterGoogle Scholar
- Esbensen KH (2002) Multivariate data analysis in practice. CAMO Process AS, OsloGoogle Scholar
- Naes T, Isaksson T, Fearn T, Davies T (2002) A user-friendly guide to multivariate calibration and classification. NIR Publications, Chichester 420 pGoogle Scholar
- Otto M (1999) Chemometrics: statistics and computer application in analytical chemistry. Wiley-VCH 314 pGoogle Scholar