Improve clinical care and patient outcomes, reduce fraud, and manage financial risk.Learn How
Elder Research partnered with the U.S. Postal Service Office of Inspector General to develop and deploy a custom solution to identify and prioritize questionable contracts and healthcare claims for investigation. Leads generated were 74% actionable, resulting in over $11 million in recoveries, restitutions, and cost avoidance in the first year.Download PDF
Elder Research developed a provider risk scoring model that enabled targeted intervention with low quality providers and reduced per patient cost by nearly 20 percent.Download PDF
By applying advanced techniques for modeling and visualizing customer records, Elder Research created a combined data and text mining solution to increase marketing efficiency and reduce churn. The model improved targeted messages which resulted in higher profitability for nTelos, a regional mobile phone carrier.Download PDF
Senior Data Scientist Robert Pitney's article "Analytics Assessment: A Blueprint for Effective Analytics Programs" was published in Predictive Analytics Times. The article discusses the importance of doing an Analytics Assessment to developing a well-conceived analytics strategy to achieve a sustainable competitive advantage and generate significant return on investment using analytics.
Data Scientist Tom Shafer has been declared the winner of its Blackmarker ML Modeling Contest. Tom’s model outperformed all other contest entries and was validated with a 250 document out-of-sample evaluation set which produced an F-measure that exceeds the threshold for a doubling of the prize money.
According to Zach Buckner, CEO of Blackmarker, “Tom’s model leveraged innovative features that he developed over top of the original data, including some NLP-based features. It also employed some clever tactics, such as altering the shape of the page targets that were provided by a commercial OCR package and by our proprietary Glyphic processing engine. This result is a testament to his exceptional talent as a data scientist. We’re delighted with his work and with the results of the contest, which will be incorporated into our application to drive even greater accuracy.”
This eBook includes Chapter 3 from industry experts Jeff Deal and Gerhard Pilcher’s book Mining Your Own Business, A Primer for Executives on Understanding and Employing Data Mining and Predictive Analytics. Chapter 3 titled “Leading a Data Analytics Initiative” covers the key challenges and considerations for business leaders employing analytics to provide data-drive insight.
Every technical project involves some sort of analytics, ranging from simply reporting key facts, to predicting new events. In this eBook we define ten increasingly sophisticated levels of analytics so that teams can assess where they stand and to what they aspire. The eBook clarifies definitions of three types of analytic inquiry and four categories of modeling technology and illustrates these levels with examples using tabular data representations commonly found in spreadsheets and single database tables. Additionally, the Levels are extended to encompass emerging data types such as time series, spatial data, and graph data, by providing data complexity as second dimension for categorization alongside algorithmic sophistication.
In two decades of mining data from diverse fields, we have made many mistakes, which may yet lead to wisdom. In this eBook, we briefly describe, and illustrate from examples, what we believe are the “Top 10” mistakes of data mining, in terms of frequency and seriousness. Most are basic, though a few are subtle. All have, when undetected, left analysts worse off than if they’d never looked at their data.
Gerhard Pilcher discussed the gap between the promise of analytics to transform a company and the actual results, and how adaptability and intuition by leadership can help close that gap.
Target Shuffling is a process for testing the statistical accuracy of data mining results. It is particularly useful for identifying false positives, or when two events or variables occurring together are perceived to have a cause-and-effect relationship, as opposed to a coincidental one.