Data Science: Unsupervised Learning
Analyst, BI/Analytics Practitioner, Data Scientist
In unsupervised, or undirected learning, you don’t have a target variable. You are just fishing in the mud, trying to find some useful general patterns in your data. Market basket analysis is a classic example of unsupervised learning. You try to find which products are purchased together, the association rules. With clustering, you try to find groups of cases, for example typical groups of customers. The most popular clustering methods include hierarchical and K-means clustering. A more advanced algorithm is support vector machines. This one can be also used for supervised learning. This session introduces all of the algorithms mentioned, theoretically and with a lot of demos in R and Python.
Familiarity with data science concepts. Knowledge of Python and / or R is helpful as well.