Tuesday, November 4, 2014

PREDICTIVE ANALYTICS


http://www.mindtree.com/sites/default/files/mindtree-thought-posts-white-paper-enabling-predictive-analysis-in-service-oriented-bpm-solutions.pdf

http://acharyavivek.wordpress.com/2013/02/20/intelligent-bpm-oracle-ibpm/

http://www.oracle.com/us/technologies/bpm/wp-intelligent-bpm-2280473.pdf

Oracle R Enterprise

http://www.oracle.com/technetwork/database/database-technologies/r/r-enterprise/overview/index.html

Oracle R Enterprise, a component of the Oracle Advanced Analytics Option, makes the open source R statistical programming language and environment ready for the enterprise and big data. 



Learn More about Oracle R Enterprise
white papers
tutorials
blogs


http://www.oracle.com/technetwork/database/database-technologies/r/r-enterprise/learnmore/index.html


R Technologies from Oracle

http://www.oracle.com/technetwork/database/database-technologies/r/r-technologies/r-offerings-1566363.html


https://blogs.oracle.com/R/entry/oracle_r_enterprise_tutorial_series

Oracle R Enterprise Tutorial Series on Oracle Learning Library

Oracle Server Technologies Curriculum has just released the Oracle R Enterprise Tutorial Series, which is publicly available on Oracle Learning Library (OLL). This 8 part interactive lecture series with review sessions covers Oracle R Enterprise 1.1 and an introduction to Oracle R Connector for Hadoop 1.1:
  • Introducing Oracle R Enterprise
  • Getting Started with ORE
  • R Language Basics
  • Producing Graphs in R
  • The ORE Transparency Layer
  • ORE Embedded R Scripts: R Interface
  • ORE Embedded R Scripts: SQL Interface
  • Using the Oracle R Connector for Hadoop 
https://stbeehive.oracle.com/teamcollab/wiki/BEAM+BI+12c+Workspace:Predictive+Analytics+Main+Page

http://www.inside-r.org/category/packagetags/machinelearning
http://en.wikipedia.org/wiki/Non-negative_matrix_factorization

terminology

Generalized Additive Model residuals
Extract Model Fitted Values

Fitting Generalized Linear Models
devianceup to a constant, minus twice the maximized log-likelihood. Where sensible, the constant is chosen so that a saturated model has deviance zero.

Oracle Nimbula: