SAS | The Power to Know
SAS Users orange radiance art

 

Best Practices in Machine Learning Applications
 

Thursday,  Nov. 3  |  4 – 5 p.m. ET

IFSUG logo

This complimentary webinar is hosted by the Insurance and Finance SAS Users Group (IFSUG).

There are common pitfalls and mistakes that every machine learning practitioner finds troubling. Building representative machine learning models that generalize well on future data requires careful consideration both of the data at hand and of assumptions about the various available training algorithms. Data is rarely in an ideal form that enables algorithms to train effectively.

During this hour, a foremost machine learning expert at SAS will provide guidance and insights on building effective models from powerful algorithms. Topics include:

  • Handling high-cardinality variables: by-group numeric encodings, factorization machines, entity embedding and neural networks.
  • Ensemble models: bagging, boosting, stacking and other blending methods.
  • Model interpretation: surrogate models, LIME and partial dependency plots.

About the Presenter

Patrick Hall

Patrick Hall, Senior Data Scientist, SAS

Patrick Hall is a Senior Data Scientist at SAS and an adjunct professor in the Department of Decision Sciences at George Washington University. He provides guidance for SAS and its flagship customers on deriving substantive value from machine learning technologies and designs new data mining and machine learning approaches, focusing on neural networks and clustering. He holds multiple patents in automated market segmentation. Hall is the 11th person worldwide to become a Cloudera certified data scientist. He studied computational chemistry at the University of Illinois before graduating from the Institute for Advanced Analytics at North Carolina State University.


About the Insurance and Finance SAS® Users Group (IFSUG)
The Insurance and Finance SAS Users Group is open to all SAS users in the insurance and finance industry. Its purpose is to provide a forum for the exchange of ideas, information and best practices with SAS solutions. To learn more or to join our group, visit the IFSUG website.

Registration Now Closed -- All Available Seats Are Filled

 
*
*
*
*
 
*
 
*
 Under 18
 18 - 24
 25 - 34
 35 - 44
 45 - 54
 55 - 64
 65 and older
 Prefer not to answer
*
 Fewer than two
 Two to five
 Five to 10
 10 to 15
 More than 15
*
 Yes
 No
 
 

Back to Top