DATA MINING ONSITE VISIT/DEMONSTRATIONS
Locations: PHILADELPHIA, SEATTLE and SAN DIEGO
Dates: August 10-11 (Seattle), August 14-18 (San Diego), August 24-25 (Philadelphia)
Dr. Dan Steinberg, CEO of Salford Systems, will be visiting Seattle, San Diego and Philadelphia in August. If you and your colleagues would like to arrange a meeting time at your offices, please let us know as soon as possible. Dr. Steinberg can discuss your specific application and how these would benefit from predictive modeling and data mining. In demonstating the power of data mining and our tools, he can, if you wish, use your data. See more information regarding Salford Systems at the bottom of this email.
**
DATA MINING TRAINING
Locations: PHILADELPHIA and SEATTLE
Dates: August 9 (Seattle) and August 14, 15, 16, 17, 18 (Philadelphia)
SEATTLE DATA MINING TRAINING
Sponsored by the American Statistical Association
Wednesday, August 9th
Computer Technology Workshops: CE_34T, CE_37T, CE_40T
http://www.amstat.org/meetings/jsm/2006/index.cfm?fuseaction=registration
HANDS-ON DATA MINING TRAINING IN PHILADELPHIA
August 14-18, 2006
Register for 1, 2, 3, 4 or 5 days.
http://www.salford-systems.com/servicestraining.php
Please let me know if I can help you with any of the above.
Best regards,
Lisa Solomon
Ph: (619)543-8880
Email: lisas@salford-systems.com
More information about Salford Systems Data Mining Tools:
Salford Systems has focused its efforts on becoming best of breed in the specific areas of decision, trees, boosted decision trees, spline regression (MARS) and tree ensembles (bagging, arcing, random forests). Our tools are based on the original code of the creators of the proprietary CART algorithm. CART's creators continue to collaborate with Salford Systems to refine CART and to develop the next generation of data-mining tools. We have extended the original code in collaboration with Jerome Friedman and Leo Breiman to include innovative features not found elsewhere, including patent pending extensions. We have exclusive access to the proprietary source code written by Jerome Friedman for CART, TreeNet, MARS, PRIM, and will be adding further Friedman tools in 2007.
* Robustness: Our software is known for its capability of handling large problems, and of running problem free no matter how complex the sequence of analytical steps required in a user session.
* Our latest versions include modeling automation to radically accelerate model development. In a recent client visit one of our senior scientists was able to build a response model in one day that significantly outperformed the client's model that had been hand crafted over 3 months. The built-in modeling automation allowed him to find that model in hours.
* Ease of use: Our unique visual displays make it easy for users to understand the data and quickly move to effective models. It is possible to start building a model in as few as 3 mouse clicks (open a data source, select a target variable, go).
* Analytical power: Our tools have now been used to win 14 data mining "medals" since 2000.
* Availability on multiple platforms
* Command-line support allows automated runs.
This archive was generated by hypermail 2b30 : Wed Jul 12 2006 - 20:50:36 MEST