Public Service Delivery & Citizen Experience Dataset

This is a synthetic dataset generated from high-quality expert-labelled seed data. All records are algorithmically derived — statistical distributions, inter-field correlations, and annotation characteristics faithfully replicate real-world patterns from the source data, while ensuring no real individual, organisation, or transaction can be identified or reconstructed.

The African Public Service Delivery & Citizen Experience Dataset aggregates 300K+ citizen service interaction records from government agencies across Nigeria, Kenya, Rwanda, and Senegal — four countries at different stages of digital government transformation. Rwanda's Irembo e-government platform and Kenya's eCitizen portal contribute structured digital service records; Nigeria and Senegal contribute a mix of digital service logs, call-centre records, and field survey data from citizens who interacted with public agencies in person.

Each record captures the service request type from a 28-class taxonomy (passport/ID, business registration, land title, tax filing, utility connection, social protection enrolment, health certificate, court filing, and more), the channel used (online portal, USSD, call centre, in-person), the number of interactions required for resolution, total resolution time in days, the final outcome (resolved, partially resolved, escalated, abandoned), and a citizen satisfaction score (1–5 stars) where collected. Demographic proxies (gender, age group, urban/rural) are included for equity analysis.

The dataset is designed for GovTech AI teams building service-routing intelligence (matching citizen requests to the right agency and officer), resolution-time prediction, proactive escalation triggers, and citizen satisfaction drivers analysis. It also supports policy evaluation — measuring the impact of digital channel rollout on resolution rates and satisfaction across demographic subgroups — and equity auditing of public service access and quality.