Join us for three days of the best in Machine Learning, Spatial, and Analytics on the web.
September 13-14, 2022
Level up your knowledge and watch the leaders in Machine Learning, Spatial, and Analytics work their magic. Ask questions, get real answers and watch the solutions, all from the comfort of your desk. Attend one session or all three days! We look forward to seeing you!
Spring TechCast Days
If you wish to attend multiple days, you must register for each day separately.
May 24th
Analytics+
Demo Day
HOSTS
Edelweiss Kammermann &
Dan Vlamis
*Session start times are subject to change.
11:05am–11:50am ET
OAC: ML and AI Features and Integration Success
Edelweiss Kammermann
11:50am–12:30pm ET
Strategies and Layouts for Data Visualization
Dan Vlamis & Tim Vlamis, Software Solutions
12:30pm
Break
12:35pm–1:05pm ET
“Automating Repetitive Tasks in OAC with RPA”
KJ Fenton, UiPath
1:05pm
Break
1:10pm
Real-World Lessons from Essbase 21c On-Premises Deployments
Tim German, Quibix
2:00pm ET
Closing Remarks
May 25th
Graph
Demo Day
HOSTS
Melli Annamalai & Roger Cressy
*Session start times are subject to change.
11:00am–11:05am ET
Welcome
Roger Cressey, Melli Annamalai
11:05am-11:55am ET
Build RDF Graph SPARQL Endpoints with Autonomous Database
Royta Yamanaka, Oracle & Ramu Maurakami Gutierrez, Oracle
11:55am–12:00pm ET
Break
12:00pm–12:55pm ET
Grafting Grifters: Identify and Display Patterns of Corruption with Oracle Graph Studio
Jim Czuprynski, Zero Defect Computing, Inc.
12:55pm-1:05pm ET
UiPath Brief Demo
KJ Fenton, UiPath
1:05pm–1:10pm ET
Break
1:10pm-1:35pm ET
Analyzing Report Usage with Graph Studio: How Graph Provides a Different Perspective
Mark Daynes, Beyond Systems
1:35pm–2:00pm ET
What’s New in Oracle Graph
Speaker: Melli Annamalai, Oracle
May 26th
Machine Learning Demo Day
HOSTS
Edelweiss Kammermann &
Roger Cressey
*Session start times are subject to change.
11:05am–11:50am ET
Plano Orthopedics Revenue Cycle and ML in Revenue Cycle Management Processing
Steve Chamberlin, Sensa Analytics
11:50am–12:35pm ET
Oracle Marketing Data Science Tips & Best Practices
Joe Darschewski, Oracle
12:35pm-12:45pm ET
Using Oracle Machine Learning for RPA
KJ Fenton, UiPath
12:45pm–12:50pm ET
Break
12:50pm-1:20pm ET
Creating Unbiased TV Commercial Exposure Estimate
Matthew Pocernich, Oracle
1:20pm-2:00pm ET
Learn How to Invoke Your Python Functions from SQL with OML4Py
Mark Hornick, Oracle
Analytics Abstracts
Edelweiss Kammerman
Oracle Analysis Cloud includes many out-of-the-box Ml And AI features that make the life of the business analysts easier. These are very intuitive features that bring a lot of value and reduce the time of data cleansing and data analysis. But there is more. In the latest releases, we can find more tight integration with Oracle Machine Learning as well as with OCI AI Services. In this session, you will see all these features and how to integrate with ML and AI in detail and with live demos.
Dan Vlamis & Tim Vlamis
Vlamis Software Solutions
Enterprise data is complex and knowing how to design interactive and effective presentations requires more than the basics. This class will review specific combinations of graphs and visualizations designed to reveal complex multi-dimensional insights, patterns, and data relationships. You’ll see how to use scatter plots, trellis graphs, parallel coordinate graphs, and pivot heat maps to reveal patterns and relationships otherwise hidden in multi-dimensional data sets. You’ll learn three fundamental layouts and optional techniques so that you always have a strategy to build a comprehensive presentation. Just as mother nature has certain patterns that are used over and over again, you’ll see how to construct an effective set of visualizations, interactions, and drill paths using patterned layouts and data visualization best practices.
Enterprise data is complex and knowing how to design interactive and effective presentations requires more than the basics. This class will review specific combinations of graphs and visualizations designed to reveal complex multi-dimensional insights, patterns, and data relationships. You’ll see how to use scatter plots, trellis graphs, parallel coordinate graphs, and pivot heat maps to reveal patterns and relationships otherwise hidden in multi-dimensional data sets. You’ll learn three fundamental layouts and optional techniques so that you always have a strategy to build a comprehensive presentation. Just as mother nature has certain patterns that are used over and over again, you’ll see how to construct an effective set of visualizations, interactions, and drill paths using patterned layouts and data visualization best practices.
Enterprise data is complex and knowing how to design interactive and effective presentations requires more than the basics. This class will review specific combinations of graphs and visualizations designed to reveal complex multi-dimensional insights, patterns, and data relationships. You’ll see how to use scatter plots, trellis graphs, parallel coordinate graphs, and pivot heat maps to reveal patterns and relationships otherwise hidden in multi-dimensional data sets. You’ll learn three fundamental layouts and optional techniques so that you always have a strategy to build a comprehensive presentation. Just as mother nature has certain patterns that are used over and over again, you’ll see how to construct an effective set of visualizations, interactions, and drill paths using patterned layouts and data visualization best practices.
Tim German, Qubix
With the 21c release, customers have the option to run the same version of Essbase that runs in the Oracle Cloud – along with all its formerly “cloud only” features – on on-premises servers or those of other cloud infrastructure providers. But there are many subtle and not-so-subtle differences between an “independent deployment” of Essbase 21c and previous on-prem versions, spanning interfaces, integrations, automation, architecture and infrastructure. In this session, we’ll talk about real-world on-premises implementations of both Linux and Windows versions of 21c.
Graph
Abstracts
Abstracts
Ryota Yamanaka, Oracle & Ramu Murakami Gutierrez, Oracle
Resource Description Framework (or RDF) Graphs are a powerful tool for knowledge graphs and linked data applications, used in a variety of industries such as public sector, finance, life sciences, and more. RDF enables exchange of data and metadata on the Web. SPARQL is the standard query language for RDF and many organizations provide SPARQL endpoints to support queries over HTTP, making it possible to link data sources. We will show how you can develop secure SPARQL endpoints and publish your datasets using the Always Free Services of Oracle Cloud. Such a public endpoint will enable you to connect RDF graphs in Autonomous Database to the “Web of Data.” We will also demonstrate how you can get started with RDF graphs in minutes when using the Graph Studio feature of Autonomous Database.
Jim Czuprynski, Zero Defect Computing, Inc.
Ever wonder how your bank knows when to text you when a potentially fraudulent transaction appears on its radar? It’s simple if you know how to apply the right algorithms to financial transactions so that outliers become readily apparent. These techniques can be applied to just about any anomalous behavior, and property graph technology helps make outliers stand out dramatically. Through presentations and online demonstrations, this session will:
Show how to use appropriate ML algorithms already built into Oracle Database to uncover anomalous patterns
Demonstrate Oracle Graph Studio’s capabilities to identify and display outliers more clearly
Introduce the basic concepts of PGQL (Property Graph Query Language) to look at tabular data in completely new ways
Mark Daynes, Beyond Systems
Investigating report and analytics execution, performance and the users that are running them can be done in a number of ways using BI and analytics tools, but one that brings a fresh perspective is Oracle Graph Studio. Using Notebooks, we can investigate the relationships between the users and the reports that are being executed and use that as a basis for exploration to find insights you may not have seen previously. In this session, we will discuss the process to do this and see some insights in action.
Melli Annamalai, Oracle
Learn about the latest features in Oracle Graph technologies, including Graph Studio and Server and Client.
Machine Learning Abstracts
Steve Chamberlin, Sensa Analytics
Healthcare
organizations are faced with the challenge of efficiently processing thousands
of claims each day. There two primary reasons submitting claims has become
difficult:
1.Human interaction for submitting and reviewing claims will cause errors due to the increased volume of requirements for the procedures.
2. The requirements for submitting claims change daily when looking at the broad spectrum of healthcare plans and
changing insurance company policies.
A more effective method must be put in place to help the Revenue Cycle Management (RCM) team. They need a smarter way to avoid mistakes when coding a claim and track denied claims for a quicker response.
Plano Orthopedics Sports Medicine Center (POSMC) has teamed with Sensa Analytics to use Oracle’s
Machine Learning in this process. Machine Learning is used to search the setup or coding of claims for errors prior to submitting them to the insurance
companies. This means faster reimbursement of the claims. It can mean the difference of getting paid in 2 weeks instead of 2-3 months. This can be
anywhere from a 2-6% improvement which adds up hundreds of thousands annually.
Another aspect of ML is being used to find denied claims and categorize them. This is a monumental
task for the RCM team. An individual in RCM could have over 500 claims to manage every day. This means creating or correcting a claim in less than 1
minute continuously throughout the day every day. This makes it impossible to manually review multiple reports and systems to find claim issues and update them with the proper response.
Oracle’s ML and Sensa have helped to identify over $600,000 in additional claims revenue in a 3-month span.
Learn how Sensa applied ML to the data, the technical architecture and process. Examples for how ML was applied will be discussed in the session. The use cases will cover the goals, end results and lessons learned from the project.
Joe Darschewski, Oracle
Oracle Marketing Data Science team made the transition from Oracle R Enterprise to OML on Oracle Cloud Infrastructure. Transforming our scripts has provided us a unique opportunity to revisit how we model, use of the script repository, and logging throughout our process. We present our best lessons learned with examples including using oml.do_eval for performance tuning, using cx_Oracle for embedded sql scripting within Python blocks, tracking time to execute through programs and recording step by step logs for tracking performance and completion of jobs. Etc.
Mark Hornick, Oracle
Oracle Machine Learning for Python allows users to run user-defined Python functions in Python engines spawned and managed by the database environment – from Python (of course), but also SQL. The SQL interface makes it easy for applications and dashboards to deploy Python-based solutions in applications and dashboards. This feature, called Embedded Python Execution (EPE), also allows you to take advantage of system-provided data-parallelism for those “embarrassingly parallel” use cases. Supporting OML4Py EPE is the ability to store user-defined Python functions in the database script repository as well as Python objects in the database datastore – no more separate flat files to manage, back up, and secure. Join us to learn about this powerful capability.
Matthew Pocernich, Oracle
To measure viewership for TV commercials, information is gathered from millions of TVs using internal software that shares viewing information through the internet. TV viewing data may be joined with household demographic information such as income, age, education and presence of children to describe the reached population. Since connected TVs are not distributed randomly, un-adjusted reach estimates are biased. To get unbiased national estimates of commercial viewing, it a standard practice in the ad-tech industry is to use weighting to adjust for biases. This talk discusses the way the R is used to evaluate initial biases viewing data, select relevant attributes, apply a raking algorithm to estimates device weights and evaluate the robustness of the resulting weights. Raking algorithms are used when information is only available for the marginal distributions. On a daily basis, tens of millions of TVs need to be weighted, dozens of attributes considered each with from 2 to hundreds of levels. Consequently, the efficiency of the methodology is important. The implementation of this methodology uses Oracle Machine Learning for R (OML4R) which provides such efficiencies.