Analytics and Data Summit 2018

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. Designed for problems involving both large and small volumes of data, Oracle R Enterprise integrates R with Oracle Database, enabling R users to run R commands and scripts for statistical, machine learning, and graphical analyses on database tables. In this session, we will briefly introduce Oracle R Enterprise, but focus on the new features of Oracle R Enterprise 1.5.1, including:

• OREdplyr - transparncy layer package based on the popular and powerful dplyr package from the tidyverse
• OAAgraph - enabling combining Oracle Spatial and Graph PGX with OAA machine learning from a common R interface
• Explicit Semantic Analysis algorithm - new text analytics capability to extract features based on human curated topics, such as from Wikipedia
• Partititioned models - automated building and scoring of predictive models based on  data partitions
• and more...

Join us to see what how you could take your R-based analytics to the next level of functionality, scalability, and performance.

Graph analysis is an effective methodology in data analysis which considers fine-grained relationships between data entities. This session presents an example of anomaly detection using graph analysis on data stored in Oracle Database with Oracle R Enterprise and the OAAgraph package, using a real-world public data set.
Specifically, we analyze the public information of medical transactions from CMS (United States Center for Medicare and Medicaid Services) for the year 2012. We load this data set as a graph into Oracle Database, and use the OAAgraph package to apply graph analysis algorithms whose results are be visualized using R's ggplot2 visualization package. 
Our analysis successfully identifies suspicious records in the data set -- medical providers who perform procedures far from their specialties. We also discuss techniques how to reduce false positive outcomes.