Links to Tutorials on Data Analysis

Gentle Introductions:

Chemometrics Column in J. Chemometrics by Richard Brereton introduces "Points, vectors, linear independence and some introductory linear algebra".

Applied Maths:

These videos are being used to advertise their data analysis software (BioNumerics) and hence are specific to that software. This emphasis is very much on the software rather than a description of the techniques being used. Description of how to do principle Component Analysis and Discriminant Analysis using their software.

The same company has some videos on peak matching and statistical analysis of spectral data. 

Department of Food Science, University of Copenhagen:

A full set of chemometrics lessons including  two series of videos on PCA and PLS. Although the videos are relatively long the description of the theory of the method is good and there is good coverage of related topics such as what to do with outliers in PCA. Also videos on spectral pre-processing, variable selection and co-clustering.

The same organization has a paper available titled “Multi-block Methods for Exploratory Data Mining in Food Technology” at:-

Victor Lavrenco (University of Edinburgh):

Sets of short videos on a variety of topics available form

Don’t be put off by the video that immediately starts up when you get to this page. Most of the topics below are much easier to understand.

Topics recommended (though I haven’t looked at every video in the series)

Thinking about data (Under Applied Machine Learning:-a set of 25 short videos, not all of them relevant to our work)

Some of them quite basic, but some are quite good. The last video in the series (Detect Outliers by Visualizing the Data) emphasizes the need to plot data to assess outliers and how trying to filter out outliers using acceptable ranges can often fail.

Specific Topics that are relevant to our work (each video is generally less than 5 minutes)

PCA-Data Analysis (12 videos)
K-means clustering-Data Analysis (9 videos)
Hierarchical Clustering (5 videos)
Decision Tree –Machine Learning (8 videos)
Nearest Neighbour Methods – Machine Learning (10 videos)
Mixture Models_Data Analysis (5 videos). This is a probabilistic method of soft clustering.

Last updated: August 26, 2016 at 8:02 am