Contributing Disciplines

• Databases. Indispensable if data mining is to be efficient on extremely large datasets (Gigabytes to Terabytes). A lot of data mining work is on smaller, in-core datasets, however.
• Statistics. Basic algorithms for data analysis, mostly numeric, like regression or statistical clustering.
• Machine Learning: Algorithms for various analysis tasks involving non-numeric data, application background knowledge, complex hypotheses.

  Þ Statistics and Machine Learning provide the "toolbox" of analysis methods to be used in a data mining application but are less concerned with data storage and data preparation tasks (even though in practice, every statistics and ML application always features these steps!)

---