/Sensory-and-Chemometrics-Data-Science-Seminar-Series

Two-part seminar series introducing a basic quantitative and computational toolkit for analyzing data acquired from experiments in sensory science and chemistry. View the notebooks via NBViewer.

Primary LanguageJupyter Notebook

Sensory and Chemometrics Data Science Seminar Series

This two-part seminar series aims to introduce a basic quantitative and computational toolkit required to effectively explore and analyse data acquired from experiments in sensory science and chemistry.

Although the language of instruction will be R, no prior programming experience is assumed. The seminars will introduce some basic programming concepts in the R language and illustrate those basics through statistical analyses on real datasets. There will also be discussions on methods to interpret the results of such analyses.

Everyone is welcome and encouraged to attend, especially those who participated in the Pandan sensory panels held in January 2018. Reference samples of the flavours used in the January 2018 panel will be provided again for the benefit of those who were not able to participate in the panel.

The seminars will be conducted jointly by Augustine Koh, from VK Creative Aromatics International Pte Ltd, and Aaron Thong, from the A*STAR Biotransformation Innovation Platform.

The outline of the seminar series is as follows:

Session 1: Sensory with R – panel testing and QDA

  • Analysis of the January 2018 Pandan flavour panel
    • How to judge panel performance
    • Testing of consensus
    • Testing of repeatability
    • Judging of product effect
    • Building sensory profiles
  • Background of sensory statistical analysis
    • Sensory with R – an introduction
    • ANOVA testing of panellist performance
    • A statistical look at Quantitative Descriptor Analysis (QDA)

Session 2: Chemometrics with R

  • What is Chemometrics?
    • Extracting information from chemical systems by data driven means
    • Descriptive applications – understanding the underlying relationships and structure of the system
    • Predictive applications – modelling of the data in order to predict new properties or behaviours
  • Multivariate Analysis
    • Dealing with large datasets and the need for dimensionality reduction
    • Introduction to Principal Components Analysis (PCA)
    • PCA with R

Seminar Series Schedule

Session Date and Time Location
1 15 March 2018 at 4:45pm A*STAR Biotransformation Innovation Platform
2 TBC A*STAR Biotransformation Innovation Platform