evidoria

← Volver al listado

Buena práctica Importada

ClassifAI y Survey Assist — el pipeline de IA de la ONS británica que redujo a la mitad el tiempo de codificación de ocupaciones

Reino Unido · Newport · Ver el perfil de Reino Unido

Evidencia: Observacional / pre-post Top 7% 87/100 · Pregunte a Evidence Copilot sobre esta práctica

La ONS británica creó ClassifAI, código abierto que codifica texto de encuestas en clasificaciones oficiales, reduciendo a la mitad el tiempo de codificación de ocupaciones en una gran encuesta laboral, y ahora pilota Survey Assist bajo registro público de transparencia.

-50% %
Reduction in occupation-coding processing time on ASHE
~90,000 households
Households potentially covered by Survey Assist once deployed (annual, once deployed)
~2.25 million instances
Survey instances Survey Assist could cover annually (annual, once deployed)
ClassifAI y Survey Assist — el pipeline de IA de la ONS británica que redujo a la mitad el tiempo de codificación de ocupaciones

Detalles

Madurez
En expansión
Promotor
UK Office for National Statistics (ONS) - Data Science Campus
Período
2023-present
Palabras clave
official statistics, government data science, labour statistics

Contexto

National statistics offices spend significant staff time manually coding survey respondents' free-text answers into standard classifications such as the Standard Occupational Classification (SOC) and Standard Industrial Classification (SIC), a bottleneck as survey volumes grow. The UK's Office for National Statistics (ONS) Data Science Campus built ClassifAI, an open-source Python package that combines semantic search over previously coded examples with retrieval-augmented generation.

Objetivos

ClassifAI aims to assign free text automatically to classifications including SOC, SIC, COICOP and their international equivalents (ISCO, ISIC). Building on the same pipeline, the ONS is separately piloting 'Survey Assist' on the Transformed Labour Force Survey, using AI-generated follow-up questions to sharpen SIC/SOC coding.

Actividades

ClassifAI, according to the ONS's own account, is now coding labour market data in production, including a large-language-model application used on the Annual Survey of Hours and Earnings (ASHE). Survey Assist, which could cover around 90,000 households (about 2.25 million survey instances a year) once deployed, was published on 30 October 2025 under the UK's Algorithmic Transparency Record Standard (v3.0), documenting its pilot status, data flows, human clerical review and a planned evaluation against gold-standard coded data.

Resultados

The ONS's own account states that the large-language-model application built on ClassifAI halved processing time for occupation coding on ASHE. Survey Assist, the newer follow-up-question tool, has not yet published accuracy or consistency results, as testing was still under way through 2025.

Conclusiones

The Survey Assist Transparency Record sets a materially higher public-transparency bar than most comparable AI deployments in this dataset, but the well-evidenced part of the impact claim remains limited to the ASHE processing-time figure, with Survey Assist itself still unproven in production.

Implementación
Coste indicativo
Medio (50 000–500 000 €)
Tiempo hasta resultados
Medio (1–3 años)
Personal y competencias
ONS Data Science Campus (in-house team, Newport)

Condiciones para el éxito

  • Open-source (MIT-licensed) codebase enabling reuse and scrutiny
  • Retrieval-augmented generation over a library of previously coded examples
  • Publication under the UK Algorithmic Transparency Record Standard (v3.0)
  • Human clerical review retained alongside the AI-generated codes

Modos de fallo comunes

  • Survey Assist has no published accuracy or consistency results yet; testing was still under way through 2025

Habitualmente financiado por

National / regional programmes

Vías de financiación indicativas para prácticas de este tipo: compruebe siempre las convocatorias vigentes y las reglas de elegibilidad de cada programa.

Kit de replicación

Artefactos reutilizables de esta práctica — tal como los publican sus fuentes.

¿Usted implementa esta práctica? Reivindíquela — los implementadores verificados obtienen un contacto público en la página y pueden proponer correcciones.

Fuentes de datos

De dónde se obtuvo la información de esta práctica, y cuándo.

Adjuntos

Prácticas similares que pueden serle útiles