- Applicable in many areas of chemistry.
- No comparable volume currently available.
- The field is becoming increasingly important.
Rather than overshadowing the concepts with theoretical background, this book uses application-oriented examples to illustrate how chemometrics techniques can be applied to complex scenarios with multiple and dynamic variables. The book presents a diverse selection of topics that include sampling, modeling, experimental design, calibration, pattern recognition, data analysis techniques, algorithms, and error. This second edition has been completely revised to feature new chapters on principal component analysis, self-modeling curve resolution, and multi-way analysis methods. It includes expanded material on normal distributions, sampling theory, signal processing, and digital filtering.
Embracing the growing role of chemometrics in some of the latest research trends, such as quantitative biology, bioinformatics, and proteomics, this book also identifies several areas for future development and applications. Practical Guide to Chemometrics, Second Edition continues to offer a reliable source of useful information in a style that is accessible to all levels of students, professionals, and researchers involved in analyzing scientific data.
For food science in general and food analysis and control in particular, there are several problems for which chemometrics are of utmost importance. Traceability, i.e. the possibility of verifying the animal/botanical, geographical and/or productive origin of a foodstuff, is, for instance, one area where the use of chemometric techniques is not only recommended but essential: indeed, at present no specific chemical and/or physico-chemical markers have been identified that can be univocally linked to the origin of a foodstuff and the only way of obtaining reliable traceability is by means of multivariate classification applied to experimental fingerprinting results.
Another area where chemometrics is of particular importance is in building the bridge between consumer preferences, sensory attributes and molecular profiling of food: by identifying latent structures among the data tables, bilinear modeling techniques (such as PCA, MCR, PLS and its various evolutions) can provide an interpretable and reliable connection among these domains. Other problems include process control and monitoring, the possibility of using RGB or hyperspectral imaging techniques to nondestructively check food quality, calibration of multidimensional or hyphenated instruments etc.