projects

Enhancing text analytics and capacity building of youth at risk

Completed
A group of people sit facing a presenter in a UNHCR vest
Start Date
Total Project Cost
USD 27,798.000
Country
El Salvador
Project Team
UNHCR El Salvador , Municipality of San Salvador

Challenge

UNHCR El Salvador collects semi-structured qualitative data from internally displaced people and community members at risk, which is analysed manually by the operation and partners. A more systematic analytical approach could improve the identification of overarching trends and patterns, which could strengthen existing advocacy efforts and community-based protection interventions.

Solution

Develop an analytical framework to support qualitative text data analysis that would provide additional insights. The project also involves community-based participatory aspects, with opportunities for at-risk youth in El Salvador to learn important data science skills.

Impact

Better insight into overarching protection trends and drivers of protection risks, their effects on communities, and possible community-based solutions. This will facilitate advocacy efforts, support programmatic planning, and help shape effective solutions.

Project impact

122
focus groups analyzed, representing insights from 1,336 participants
400
hours of audio & 8,000+ pages of transcripts processed using AI-enabled methods
19
youth completed a 3‑month data science programme, earning Google/Coursera certificates

Other information

UNHCR El Salvador transformed the way participatory assessment data is analyzed by combining artificial intelligence, speech-to-text technology, and large language models to process one of the organization's largest qualitative datasets. The project converted more than 400 hours of focus group discussions involving 1,336 participants into structured, searchable evidence, allowing teams to rapidly identify protection concerns, barriers to education, livelihood challenges, and community priorities that would otherwise remain buried in thousands of pages of transcripts. Beyond improving decision-making, the initiative developed reusable tools for transcription, anonymization, and qualitative analysis that can be replicated across UNHCR operations. The project also invested in local talent, training youth at risk in data science and R programming, with 19 participants successfully completing the programme and contributing to the analysis of national survey and community data. The resulting approach demonstrated how generative AI can help humanitarian organizations better understand community voices at scale while maintaining data protection standards and reducing the time required for qualitative analysis. Read more about this initiative in our webstory.