An integrity office wants to explore how vendors, contracts, and employees connect when hunting fraud rings. Which NoSQL type best supports relationship-centric queries?

Fraud and integrity investigations need to walk how vendors, contracts, and employees connect—multi-hop relationship queries—so a graph database optimized for traversing entity relationships is the right NoSQL type. Edge-centric storage makes ring detection and conflict-of-interest paths natural. A key-value cache that only stores password hashes serves authentication lookups, not exploratory link analysis across civic entities. A flat file of unrelated CSV rows with no edges forces analysts to reinvent joins outside the store and scales poorly for iterative path queries. A document archive that cannot express cross-entity links may hold case notes but will not efficiently answer “who connects to whom through which contracts.” Exam caveat: graphs complement rather than replace ledgers; monetary postings may still live in relational systems while the integrity office explores links in a graph. Operational check: load a small vendor–contract–employee graph from known fixtures, run a bounded-depth traversal for a suspect vendor, and verify the path results match the paper case file before expanding production feeds. In production terms for graph traversal for fraud rings, assign an owning team, wire a dashboard that shows the pattern is healthy under peak civic load, and require a staging drill that rehearses failure and recovery before resident impact. Revisit the choice when traffic shape, retention, or integrity needs change, and keep a named rollback owner on the change record. Re-test both the happy path and the failure path for graph traversal for fraud rings after material dependency upgrades so regressions are caught before the next public release.