• 2-Day Predictive Analytics and Affinium Model Implementation Training at your site to set-up Model, access the data, orient new users to its features and layout, initiate data cleanup and visualization, introduce the basics of data mining, generate initial model and reports, interpret model results and score.  At the conclusion of the training, users will be encouraged to proceed with modeling in preparation for a follow-up Model application workshop.  The instructor will develop and refine a model in parallel for use in the second workshop.  View a detailed topic outline of the two-day Affinium Model training.
• Development of One Model^1,2 from a specific data environment.  Key assumptions of the program include:
  • Must have clean data
  • Data should be pre-processed into a flat file structure in ASCII format with data labels in the first row
  • A complete data dictionary with definitions of all eligible values should be provided and the target variable clearly identified
  • All key customer assumptions must be documented to ensure a detailed program
• 2-Day Interactive Model Application Workshop within six weeks of the initial Predictive Analytics and Affinium Model Implementation Training.  The focus and general outline for this follow-up two-day workshop will be formed at the conclusion of the initial Model training event in the context of your business objectives and resources.  Participants will have the opportunity to experience real-world model-building after the initial training, and document challenges to address when returning for the follow-up workshop.  The application workshop will involve iterating through the model-building process in hands-on fashion with Affinium Model, reinforcing and refining the model as time permits, and interpreting results.
• 10 Hours of Production Telephone and eMail Support³ to assist in production transition and interim remote requests.  This feature provides a data mining expert on a retainer basis by phone and email.  Time is be documented via timesheet and allocated in 15-minute increments.  Unused time will expire 180 days after package initiation.
   

1. Based on unknown data conditions, model results will vary. A "best effort" model will be produced within the workshop time available.

2. A completed prototype model deliverable is one which has created an initial scored output file. Accuracy of results will vary, based on previously unknown data conditions.

3. All unused services expire 180 days after package initiation. Full consumption of services is encouraged to attain maximum effect.