Case studies | Applied Insights | Real World Data Science – Real World Data Science

Our case study content is in development. Interested in contributing a case study to Real World Data Science? Check out our call for contributions.

Why 95% Of AI Projects Fail and How to Change the Odds

Drawing on two decades of work building AI systems for medicine discovery, Lee Clewley argues that the success of AI models rests on three principles.

Deploying LLMs for Nonprofits: 10 Lessons from Knowbot

The Mike Hudson Foundation has developed an LLM-powered ‘answer engine’ for non-profits. Our recent conversation with them surfaced some valuable insights for practitioners…

Deploying Agentic AI - What Worked, What Broke, and What We Learned

Reproducibility

Machine learning

AI scientist and researcher Francis Osei investigates what happens when Agentic AI systems are used in real projects, where trust and reproducibility are not optional.

Defining Purposes and Uses to Support the Development of Statistical Products in a 21^st Century Census Curated Data Enterprise Environment

Data Integration

Statistical Products

Learn about researchers’ plans to develop the Curated Data Enterprise through a use case research program.

Translating the Curated Data Model into Practice - Climate resiliency of skilled nursing facilities

Data Integration

Statistical Products

How does the Curated Data Enterprise Framework work? Take a look at our demonstration use case on the resilience of skilled nursing facilities.

Vicki Lancaster, Stephanie Shipp, Sallie Keller, Henning Mortveit, Samarth Swarup, Aaron Schroeder, and Dawen Xie
University of Virginia, Biocomplexity Institute

Advancing Data Science in Official Statistics – What is the Curated Data Enterprise?

Data Integration

Statistical Products

The US Census Bureau faces the challenge of addressing complex questions requiring novel datasets and sources to answer. Official statistical agencies and public- and…

Sallie Keller, Stephanie Shipp, Vicki Lancaster, and Joseph Salvo
University of Virginia

Advancing Data Science in Official Statistics – The Policy Problem

Data Integration

Statistical Products

In this article, we introduce a new approach for the US Census Bureau to produce statistical products using survey, administrative, procedural and opportunity data. This is…

Sallie Keller, Stephanie Shipp, Vicki Lancaster and Joseph Salvo
University of Virginia

Forecasting the Health Needs of a Changing Population

Health and wellbeing

Taking NHS usage of the 1 million people in and around Bristol, then combining with population forecasts into a modelling framework has created a sophisticated but…

Luke Shaw (BNSSG ICB), Rich Wood (BNSSG ICB, University of Bath), Christos Vasilakis (University of Bath), Zehra Onen Dumlu (University of Bath)

Deduplicating and linking large datasets using Splink

Crime and justice

Robin Linacre introduces an open source tool, developed by the UK Ministry of Justice, which uses probabilistic record linkage to improve the quality of justice system data.

Learning from failure: ‘Red flags’ in body-worn camera data

Crime and justice

When body-worn cameras were rolled out to juvenile correctional officers in Texas in 2018, senior leaders hoped proactive analysis of camera metadata could be used to…

Food for Thought: Competition and challenge design

Machine learning

Natural language processing

Health and wellbeing

From structure to setup, to metrics, results, and lessons learned, Zheyuan Zhang and Uyen Le give an overview of the design of the Food for Thought competition.

Zheyuan Zhang and Uyen Le

Food for Thought: The importance of the Purchase to Plate Suite

Machine learning

Natural language processing

Health and wellbeing

Andrea Carlson and Thea Palmer Zimmerman outline the policy issues driving the development of the Purchase to Plate Suite of data products and why linking retail scanner…

Andrea Carlson and Thea Palmer Zimmerman

The Food for Thought Challenge: Using AI to support evidence-based food and nutrition policy

Machine learning

Natural language processing

Health and wellbeing

Brian Tarran and Julia Lane introduce a collection of articles telling the story of the Food for Thought Challenge, which sought to use machine learning and natural language…

Brian Tarran and Julia Lane

Food for Thought: Third place winners – Loyola Marymount

Machine learning

Natural language processing

Health and wellbeing

Yifan Hu and Mandy Korpusik of Loyola Marymount University describe their solution to the Food for Thought challenge: binary classification with pre-trained BERT.

Yifan Hu and Mandy Korpusik

Food for Thought: First place winners – Auburn Big Data

Machine learning

Natural language processing

Health and wellbeing

Auburn University’s team of PhD students and faculty describe their winning solution to the Food for Thought challenge: random forest classifiers.

Alex Knipper, Naman Bansal, Jingyi Zheng, Wenying Li, and Shubhra Kanti Karmaker

Food for Thought: The value of competitions for confidential data

Machine learning

Natural language processing

Health and wellbeing

The Food for Thought Challenge attracted new eyes from computer science and data science to think about how to address a critical real-world data linkage problem. And, in…

Steven Bedrick, Ophir Frieder, Julia Lane, and Philip Resnik

Food for Thought: Second place winners – DeepFFTLink

Machine learning

Natural language processing

Health and wellbeing

PhD students and faculty from Worcester Polytechnic Institute and Indiana University Bloomington describe their solution to the Food for Thought challenge: an ensemble of…

Yang Wu, Aishwarya Budhkar, Kai Zhang, Xuhong Zhang, and Xiaozhong Liu

The road to reproducible research: hazards to avoid and tools to get you there safely

Reproducibility

The road to reproducibility can be a tricky one to navigate, as Davit Svanidze discovered when he set about making his bachelor’s thesis reproducible. If you are thinking of…

Davit Svanidze, Andre Python, et al.