Data Mining
Uncovering Obscure Patterns
Data mining is the process of discovering patterns in large datasets that weren't explicitly looked for. This section covers association rules, anomaly detection, clustering, and text mining — the techniques used when the question isn't fully formed yet and the data itself has to suggest the answer
Anomaly Detection
intermediateFinding the data points that don't belong — before they become a problem.
Association Rules
intermediatePeople who buy X also tend to buy Y. Finding those patterns automatically.
Clustering
intermediateGrouping similar things together when nobody told you what the groups were.
Text Mining
intermediateTurning unstructured text into numbers you can actually analyze.