Dataweave makes exploring big datasets easier for journalists
DataWeave: Deploying Human-LLM Analytics for Exploratory Structured Data Analysis
Artificial IntelligenceHuman-Computer Interaction
Summary
Exploring big collections of structured data is hard and slow for journalists because they need to understand many details and write complicated code. The authors created DataWeave, a system that helps journalists ask questions and get accurate answers from large datasets using a conversation with a language model. Unlike automatic tools that often misunderstand data, DataWeave lets users check, correct, and guide the model’s outputs as they explore. They tested it with real journalists on a complex education dataset and learned design lessons for trustworthy collaboration between humans and AI.
What this means in practice
- •For newsroom data teams: Enable reporters to interactively explore complex public datasets with guided AI assistance that handles schema and domain challenges.
- •For public sector analysts: Assist analysts in navigating frequently changing structured datasets by combining conversational queries with schema-aware AI support.
Tested on one dataset.
Authors
Raquib Bin Yousuf, Harith Laxman, Vitaliy Shkremetko, Eunice Son, Shambhavi Verma, Brian O'Leary, Venketesh Subramony, Sylvain Nazef, Jacquelyn Elias, Ron Coddington, Chris Contakes, Michael Riley, Naren Ramakrishnan
Abstract
Data journalism, the practice of using data analysis to surface newsworthy stories, depends increasingly on the ability of reporters and investigative journalists to uncover trends, disparities, and accountability narratives. In practice, exploring large structured datasets remains slow and brittle: journalists must navigate hundreds of variables across many datasets over years, understand data coding conventions, and write non-trivial analysis code while hypotheses evolve. Although LLMs are often touted as "ask in English, get SQL/answers," real newsroom workflows expose recurring failures, e.g., schema mismatches and drift, misread domain semantics and units, and silent assumptions. We present DataWeave, a system that addresses these needs by combining conversational interaction, schema grounding, analytical planning, and executable query generation to support exploratory analysis over structured data. Rather than treating LLMs as autonomous answer engines, DataWeave frames them as interactive partners whose outputs can be inspected, corrected, and steered as hypotheses shift. We present a case study with professional journalists using our system to analyze the U.S. Department of Education's Integrated Postsecondary Education Data System (IPEDS), a high-stakes public dataset with substantial domain semantics and frequent schema updates. We also report how deployment experience and iterative refinement shaped the current DataWeave architecture and its analytical workflow. Our findings distill design principles and deployment lessons for trustworthy human-LLM collaboration in structured data analysis.