From just data to relevant information

Top management could benefit a lot from Data Science. Algorithms help to find patterns where it is very difficult by humans. That is their beauty: A data sea could be converted into relevant information very quickly.

In the following example, I applied independently two algorithms, k-Means and Principal Component Analysis (PCA), to a set of 100 instances of 6 attributes. k-Means helped to identify two market segments with clearly different needs.

PCA is usually applied to dimension reduction purposes, in a data preparation step. However, in this example is surprising how clearly the first principal component (PC1) allow to classify both segments identified by k-Means (See figure). So, it could be an easy way to classify a new customer in order to offer them what best fits their needs.

Dr. Héctor Debernardo, doctordebernardo@puenteempresarial.com