EdoDiDA Chatbot

The EdoDiDA Chatbot is a natural-language interface built on top of DHIS2 for the Edo State Digital Governance and Data Management Agency (EdoDiDA). Delivered by Data Science Nigeria with Gates Foundation support, it lets non-technical government users query a DHIS2 database in plain English and get back answers with a table, a visual breakdown, and a written explanation. It is a concrete example of putting conversational AI on top of a national data platform so that data actually reaches the people making decisions.

What it does

EdoDiDA democratises data access for quick, real-world decision-making. Rather than requiring dashboards or SQL, a user asks a question in natural language and the chatbot:

• Translates the question into a query against the DHIS2 data.
• Returns the underlying data table.
• Produces a visual breakdown of the result.
• Adds a plain-language explanation and context – "snackable insights" that highlight what the numbers mean.
• Where relevant, offers recommendations (e.g. resource allocation).

Example interactions

The chatbot works across sectors covered by the state's DHIS2 data.

Ask a question, get a visual breakdown. "Plot malaria cases by LGA" returns a chart straight from the state's health data:

Ask for analysis, get plain-language insight. "How is the farming population distributed across each LGA?" returns per-LGA figures plus an explanation that flags high, moderate, and low-participation areas to guide targeted intervention:

Beyond agriculture and health, it handles cross-sector questions such as teacher allocation – returning figures per LGA plus recommendations that prioritise high-enrolment and under-resourced areas.

Why it matters

• Data democratisation – turns a specialist data platform (DHIS2) into something any policymaker can interrogate conversationally.
• Decision support – pairs raw figures with interpretation and recommendations, closing the gap between data and action.
• Grounded answers – responses are drawn from the government's own DHIS2 records, not generic knowledge.

Relevance to NAISH

EdoDiDA is a template for conversational access to government Digital Public Infrastructure. DHIS2 is the same platform the Hub's health-record digitisation use case targets, and the natural-language-over-DHIS2 pattern generalises to other national datasets. It exercises the Hub's data readiness and model lifecycle capabilities, and – run on sovereign compute – keeps sensitive state data in-country.

Built by Data Science Nigeria (DSN), supported by the Gates Foundation, on DHIS2.