Page 40 - GCN, Oct/Nov 2017

DATA, ANALYTICS AND VISUALIZATION
efforts to defraud the federal health care system are a growing problem, and they are turning to data analytics to help them more fully understand and ultimately stop the schemes.
To gain deeper insights into the data HHS collects, the OIG created a cloud- based fraud and analytics platform that provides users with centralized, acces- sible information in the form of a data visualization dashboard. The platform
is creating an environment in which employees can conduct in-depth data analysis, and it is playing an essential role in developing integrated operations to better understand and prioritize inves- tigations of waste, fraud and abuse.
HHS is using a combination of agile development, DevOps and open-source technology to make the most of its data, improve the prosecution of fraud cases and enhance the recovery of the stolen money. New features undergo multiple rounds of testing, and HHS solicits input from users and stakeholders before deploying them.
As health care systems become more complex, criminals are using more so- phisticated methods, which means HHS faces an urgent need to expand its fraud detection capabilities. Officials have noted that fraudsters frequently copy techniques from one another, so connect- ing multiple internal data sources and making them more accessible will give investigators a more holistic view of the data and help them identify trends.
In the end, the platform will help HHS curb fraudulent activity, save taxpayer money and protect federal health programs.
Finalist
A single tool for true
systems insight
Operating and Support Management Information System
Deputy Assistant Secretary of the Army for Cost and Economics, Army
The Deputy Assistant Secretary of the
Army for Cost and Economics organi- zation must track and manage more than 1,700 major weapon and materiel systems. It’s a massive undertaking that draws on widely disparate data sources, and the Army’s military readiness de- pends on DASA-CE’s ability to provide accurate and rapidly updated business intelligence.
With the Operating and Support Management Information System, the or- ganization is significantly shortening that time to insight.
OSMIS has application programming interfaces for application integration and custom visualizations, and it gives de- velopers clear structures for connecting previously siloed data and the freedom to customize their solutions for parsing and presenting that data. What were once labor-intensive static snapshots
are now quickly evolving into dynamic visualizations, developed on a common platform, that can be updated as quickly as new data is received.
The cost savings associated with no longer having to manually join and format datasets are significant, but
the bigger benefit is DASA-CE’s much improved capacity for scenario-based experimentation. Users are able to ask better questions of their data, make quicker and more informed business decisions, and optimize processes based on their findings.
By bringing state-of-the-art software and analytics techniques to its long- standing mission, DASA-CE can now focus on tackling the tough questions that analysts have struggled with for decades, such as determining the true cost of readiness in the Army.
Finalist
Transforming statewide data sharing
Self-Service Data Sharing and Visualization
Ohio Department of Administrative Services
Ohio agencies began providing data analytics outside basic agency needs for accounting and human resources
in 2015. But they quickly realized that more work was needed to facilitate the sharing of information among agen- cies, and in response, the Department of Administrative Services created a self-service data-sharing and visualiza- tion platform.
“Agencies have always been sharing data, but it was difficult because there wasn’t a good place within govern- ment to do it,” said Derek Bridges, program administrator for the Ohio Administrative Knowledge System
at DAS. “The platform allowed us to speak with agencies about how we can facilitate sharing in a way that will provide them security and privacy from an enterprise level.”
Cloudera helped DAS build the platform on top of Tableau’s visual analytics tools in the state’s private cloud environment. The platform as a service allows DAS to limit users’ ac- cess to the datasets that are relevant to their research.
Early initiatives involve collabo- rating with county health offices to compare data on performance and evaluate public health outcomes across the state. Officials are also preparing for a public procurement to determine the drivers and determinants of infant mortality rates using data from various agencies.
And Ohio’s Interactive Budget web- site enables visitors to see how money is spent in the state and how revenue is generated and allocated via the state’s budget.
“We also see applications for predic- tive and prescriptive analytics with some interesting cases and exploratory analyses in alignment with statewide initiatives,” Bridges said. “The chal- lenge for us is trying to meet people’s needs but remain anchored so we will not serve as a bottleneck for these projects and still maintain a sense of governance.”
40 GCN OCTOBER/NOVEMBER 2017 • GCN.COM