Data Mining Techniques Training

By | December 18, 2008

I recently attended a three-day training course titled “Data Mining Techniques: Theory and Practice” in Boston, Massachusetts.  The training is part of the Business Knowledge Series offered by SAS Institute Inc., which features courses authored by experts and utilizing SAS software to accomplish the objectives of the courses.  I was fortunate to have Gordon Linoff as my instructor for the class.  Gordon is a recognized expert on the subjects of SQL (Structured Query Language) and data mining, and has authored many popular books on the subject.  Unlike many instances of classroom-based training, the Data Mining Techniques class was rich with real-world business examples and discussed the evolution of processes that are now used to increase efficiency of business and solve business problems.

The course includes lectures on the following topics: data mining overview, regression, decision trees, neural networks, memory based reasoning, clustering, survival analysis, association rules, link analysis, and genetic algorithms.  Labs were included with most of the lectures, as students used SAS Enterprise Miner 5.3 to perform the analyses.  Stories and real business scenarios were scattered throughout the training to bring together the theory and the hands-on techniques.

Enterprise Miner is an incredibly powerful tool, utilizing the statistical routines that SAS is renowned for.  SAS created an easy to use interface for Enterprise Miner that shields the data analyst from the many lines of SAS code used to generate the models.  The most amazing feature of Enterprise Miner is the ability to compare the models generated by the different techniques to identify which would work the best to answer a particular question.

The Data Mining Techniques course is scheduled approximately 10 times annually across the United States.  Both Gordon Linoff or Michael Berry teach the course depending on the schedule or location.  I highly recommend the course to data analysts in any field.

Leave a Reply

Your email address will not be published. Required fields are marked *