Blog Header 1.jpg

Blog

Sort by Topic

Trends in Natural Language Processing

Stuart Price, Ph.D.

December 27, 2019

BLOG_Trends in Natural Language Processing

Deep Neural Networks (DNN) have radically changed the landscape of state-of-the-art performance in Natural Language Processing (NLP) within recent years. These versatile models are being used in many applications including text classification, language creation, question answering, image captioning, language translation, named entity recognition, and speech recognition. The state-of-the-art is changing quickly, sometimes leading to large leaps in performance with the release of new architectures. In October of 2018 Google released BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding which performed best in 11 different NLP benchmarks upon release. Since then, there have been many more models adding new components or tweaking the approach. In this article we’ll review some of the traditional machine learning methods used in deep learning and new trends such as Transfer Learning and Transformers to provide a foundation no matter what model is currently leading.

Read More

Modeling Outcomes: Explain or Predict

Peter Bruce

December 27, 2019

BLOG_Explain or Predict

A casual user of machine learning methods like CART or Naive Bayes is accustomed to evaluating a model by measuring how well it predicts new data.  When examining the output of statistical models, they are often flummoxed by the profusion of assessment metrics. Typical multiple linear regression output will contain, in addition to a distribution of errors (residuals) and root mean squared error (RMSE), values such as R-squared, adjusted R-squared, t-statistics, F-statistics, P-values, degrees of freedom, at a minimum, plus more.

Read More

Making Good Use of Negative Space in Machine Learning

Will Goodrum, Ph.D.

December 13, 2019

BLOG_Making Good Use of Negative Space in Machine Learning

Data Scientists frequently build Machine Learning models to discover interesting (rare) events in data. These events can be valuable (e.g., customer purchases), costly (e.g., fraud), or even dangerous (e.g., threat). Finding them is a “needle-in-a-haystack” challenge: the events are rare and hard to distinguish from the huge mass of overwhelmingly uninteresting cases recorded. To differentiate rare from normal events it helps to have a good understanding of normal behavior. But, how well do you actually know the haystack?

Read More

Ways Machine Learning Models Fail: Missing Causes

Mike Thurber

November 29, 2019

BLOG_Ways Machine Learning Models Fail - Missing Causes

I have identified five primary reasons why analytical models fail:

  1. Poor Organizational Support
  2. Missing Causes
  3. Model Overfit
  4. Data Problems
  5. False Beliefs

In this post, we will consider how and why missing causes in the data for training a model may result in incorrect inferences or failures.

Read More

Leveraging Data Analytics to Increase ROI

John Elder, Ph.D.

November 15, 2019

BLOG_Leveraging Data Analytics to Increase ROI-1

Reluctance to trust and rely on machine-based decisions is widespread. That is understandable; how can one be sure the automated decision system takes into account all the factors it should? Employees struggle first to learn the new technology, and then after making great progress and producing a promising model, decision-makers can still prove extremely reluctant to risk a new approach, no matter how well tests reveal its effectiveness. Still, in today’s competitive work environment, having a positive relationship with machines is essential to increasing profits and building return on investment (ROI).

Read More

Big Data and Clinical Trials in Medicine

Peter Bruce

November 1, 2019

BLOG_Big Data and Clinical Trials in Medicine

There was an interesting article in the New York Times magazine section on the role that Big Data can play in treating patients — discovering things that clinical trials are too slow, too expensive, and too blunt to find. The story was about a very particular set of lupus symptoms, and how a doctor, on a hunch, searched a large database and found that those symptoms were associated with an increased propensity for blood clots.

Read More

Detecting Hidden Fraud Risk from Public Data

Hudson Hollister

October 18, 2019

BLOG_Detecting Hidden Fraud Risk from Public Data

Detecting which of the federal government’s millions of contracts1 most likely involve fraud used to require insider access to agencies’ IT systems. Data analytics provides greater efficacy and higher hit rate than traditional investigative methods – and now can even be performed using only public data.

Read More

Be Smarter Than Your Devices: Learn About Big Data

Peter Bruce

October 4, 2019

BLOG_Be Smarter Than Your Devices-Learn About Big Data

When Apple CEO Tim Cook finally unveiled his company’s new Apple Watch in a widely-publicized rollout, most of the press coverage centered on its cost ($349 to start) and whether it would be as popular among consumers as the iPod or iMac. Nitin Indurkhya saw things differently.

Read More

The Case for Government Investment in Analytics

Jane Wiseman

September 20, 2019

BLOG_Government Investment in Analytics

Government stands to gain $1 trillion globally from using data analytics.1 Few government data teams have the resources to document their value, but those that do can show as much as eight-to-one return on their cost. There is significant non-financial benefit as well, as public faith in government may improve when saving time and money is paired with increased transparency and accountability.  

Read More

Determining Federal Grant Recipient Fraud Risk

Wayne Folta

September 6, 2019

BLOG_Determining Federal Grant Recipient Fraud Risk

Elder Research partnered with Excella Consulting to build an end-to-end grant risk estimation solution in the client’s AWS cloud. It used text mining and document classification to extract CPA Findings from audit reports and assign risk scores to federal grant recipients.

Read More

Better NLP Models: OpenAI GPT-2

Peter Bruce

August 23, 2019

BLOG_Better NLP Models - OpenAI GPT-2

I’ve been told that, in conversation, I jump in and finish other people’s sentences for them. Now there’s an app for that: GPT-2 released by OpenAI, founded by Elon Musk. GPT-2 is a natural language program that, given a prompt, will write (mostly) intelligible content. OpenAI's stated mission is “to ensure that artificial general intelligence (AGI) … benefits all of humanity.” Natural Language Processing (NLP) includes applications such as text classification, language creation, answering questions, language translation, and speech recognition.

Read More

Why Machine Learning Is the Coolest Science

Eric Siegel

August 9, 2019

BLOG_Why Machine Learning Is the Coolest Science

The absolutely coolest thing in science and engineering is machine learning, when computers learn from the experience encoded in data. I shall now support that hypothesis.

Read More

Investment Modeling Grounded In Data Science

John Elder, Ph.D.

July 26, 2019

BLOG_Investment Modeling Grounded In Data Science

Elder Research has solved many challenging and previously unsolved technical problems in a wide variety of fields for Government, Commercial and Investment clients, including fraud prevention, insider threat discovery, image recognition, text mining, and oil and gas discovery. But our team got its start with a hedge fund breakthrough (as described briefly in a couple of books1,2), and has remained active in that work, continuing to invent the underlying science necessary to address what is likely the hardest problem of all: accurately anticipating the enormous “ensemble model” of the markets.

Read More

The Persuasion Paradox – How Computers Optimize their Influence on You

Eric Siegel

July 5, 2019

DR. Data Show

How do computers optimize mass persuasion – for marketing, presidential campaigns, and even healthcare? And why is there actually no data that directly records influence or persuasion, considering it’s so important? And what’s the ideal technique for optimizing your dating life and for getting more people to wash their hands in public restrooms? It’s time for Dr. Data’s persuasion paradox “Groundhog Day”-inspired geeksplanation.

Persuasion modeling requires a “deep geek dive” – but it’s as important as it is fascinating.


Note: This article is based on a transcript of The Dr. Data Show episode. View the video here.


Read More

Healthcare Analytics: Exploration vs. Confirmation

Peter Bruce

June 21, 2019

BLOG_Healthcare Analytics_Exploration versus Confirmation

Perhaps the most active application of analytics and data mining is healthcare.  This week we look at one success story, the use of machine learning to predict diabetic retinopathy, one story of disappointment, the use of genetic testing in a puzzling disease, and a basic dichotomy in statistical analysis.

Read More

Data is Not Oil. It is Land.

Will Goodrum, Ph.D.

June 7, 2019

Blog_Data is Not Oil, It Is Land-1

It has become common to talk about data being the new oil. But a recent piece from WIRED magazine points out problems with this analogy. Primarily, you must extract oil for it to be valuable and that is the hard part. Framing data as oil is not illuminating for executives trying to value their data assets. Oil is valuable, marketable, and tradable. Without significant effort, data is not. Data has more in common with land that may contain oil deposits than it does with oil.

Framing data as a real asset may help executives understand its value.

Read More

Tiger vs. Jack – Asking the Right Questions

Daniel Brannock

May 24, 2019

BLOG_Tiger vs. Jack – Asking the Right Questions-1

One of the most fundamental contributions we can make as consultants is to help our clients ask the right questions of their data. We’re often asked to help solve problems that turn out to be too broadly or too narrowly defined—or are not aligned with achieving business goals or eliminating pain points.

Read More

A Deep Dive into Deep Learning

Peter Bruce

May 10, 2019

BLOG_A Deep Dive into Deep Learning

On Wednesday, March 27, the 2018 Turing Award in computing was given to Yoshua Bengio, Geoffrey Hinton and Yann LeCun for their work on deep learning. Deep learning by complex neural networks lies behind the applications that are finally bringing artificial intelligence out of the realm of science fiction into reality. Voice recognition allows you to talk to your robot devices. Image recognition is the key to self-driving cars. But what, exactly, is deep learning?

Read More

Discovering The Efficacy Of A New Drug

John Elder, Ph.D.

April 26, 2019

BLOG_Discovering The Efficacy Of A New Drug

The company had invested hundreds of millions of dollars investigating a new potential drug to treat a mental ailment, and they had zeroed in on a compound that showed promise but the compound was not passing the statistical tests required by the FDA. Elder Research was hired to examine the data and determine the drug’s viability. 

Read More

What's Brewing at Predictive Analytics World

Paul Derstine

April 12, 2019

 BLOG_Predictive Analytics World-Las Vegas

Join us in Las Vegas on June 16-20 for the largest Predictive Analytics World event to date, Mega-PAW, with seven tracks of sessions covering the commercial deployment of machine learning across industry sectors. The only conference of its kind, Predictive Analytics World delivers vendor-neutral sessions across verticals such as banking, financial services, e-commerce, entertainment, government, healthcare, manufacturing, high technology, insurance, non-profits, publishing, and retail.

Read More

Automating Network Entity Detection

Ryan McGibony

April 5, 2019

BLOG_Automating Data Pipelines and Network Entity Detection

Elder Research developed an automated data pipeline to cleanse data and feed a data visualization tool used to identify and explore document preparer network relationships. The solution enabled the client to automate significant portions of work, make data-driven decisions, prioritize resources, and gain new business value from the data.

Read More

Can Data Analytics Really Deliver 1300% ROI?

Paul Derstine

March 29, 2019

BLOG_Can Data Analytics Really Deliver 1300 ROI

With the right business problem, data, and applied techniques predictive analytics can deliver exceptional return on investment (ROI) by, for example, assessing and managing risk, detecting and preventing fraud, optimizing workflow and business processes, and prioritizing resources.  Predictive analytics provides business leaders with data-driven insight to determine which customers to contact for marketing, which tax returns to audit, which debtors to approve for increased credit limits, which patients to clinically screen, which customers are likely to leave, which persons of interest to investigate, and which equipment to inspect for impending failure.

Read More

Do We Have the Right Data?

Jeff Deal

March 22, 2019

BLOG_Do We Have the Right Data

In our experience the mistake of “waiting for perfect data” probably kills more projects than any other. Here’s a typical scenario:

The project starts out well. The management team defines the goals, calculates the potential return on investment, develops a project plan, gets a budget approved, assembles the team, and launches the project. The trouble starts with a desire to make sure that the data is in “good” condition. 

Read More

Transaction Classification Aids Credit Risk Assessment

Carlos Blancarte

March 15, 2019

BLOG_Transaction Classification Aids Credit Risk Assessment

A significant transformation is currently underway in the lending market. Banks are competing to provide lending decisions in a single day, with a vastly simplified customer experience as their primary way of growing market share. The key technology allowing this transformation is being able to accurately automate the credit decision; that is, to use advanced analytics to estimate a customer’s probability of default, affordability, and financial position to make the credit decision quickly.

Read More

RADR: A Powerful Visual Risk Analytics Tool

Victor Diloreto

March 8, 2019

BLOG_RADR-A Powerful Visual Risk Analytics Tool

The Risk Assessment Data Repository (RADR) is a powerful risk analytics platform used to enhance productivity in the investigation of fraud, waste and abuse. This server-based, data analytics product  fuses data from multiple sources, with sophisticated predictive and machine learning risk modeling, and an intuitive visual interface. RADR enables proactive identification of risk—namely fraud, waste, and abuse behaviors—and simplifies the investigative process. RADR provides visualizations for risk propensity and their related data so that managers, auditors, investigators, and analysts can easily access data on high-risk items and focus on the highest ROI cases.

Read More