United StatesPublic sector2024 – ongoingCARE USA
AI analysis of emergency preparedness survey comments
CARE reported AI analysis of emergency readiness survey comments in about half an hour
Problem
CARE collected open-ended survey comments about emergency preparedness. Staff struggled to interpret the volume of qualitative responses alongside readiness scores.
Approach
CARE worked with Valorem Reply and Microsoft to build an Azure OpenAI Service sentiment analysis tool. It tested the tool during a mid-year survey check-in and refined classifications through staff spot checks. CARE integrated selected positive and negative comments into its Power BI readiness dashboard.
Results
No published figures
Takeaway
The tool added staff explanations to quantitative readiness scores. CARE reported human-comparable classification accuracy and analysis in about half an hour; the earlier workflow, including approvals and sharing, took weeks or up to a month.
Categories
- Size
- Large enterprise
- Industry
- Public sector
- Business function
- Internal work and knowledge managementStrategy and organization
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
CARE described deploying survey sentiment analysis within its readiness dashboard. Staff checked the classifications, and CARE reported faster processing with accuracy comparable to human classification.
Full details are in the original sources.