Enterprise Knowledge Search
As its internal document archive grew, the City of Lohja needed a better way for HR support and ICT helpdesk teams to retrieve current guidance than manual keyword search, which made it hard for employees to find accurate information quickly.
Reported impact
The City of Lohja improved internal communication and information retrieval, built a chatbot that answers roughly 85% of support questions, and reduced the support workload by giving supervisors and employees faster access to up-to-date guidance.
The City of Lohja built an internal support chatbot on Oracle Autonomous AI Lakehouse and Oracle APEX to turn a growing municipal document archive into a searchable knowledge base. OCI Enterprise AI connects large language models to the source documents, extracts entities such as projects, decisions, and people, and stores those relationships in Oracle Autonomous AI Database with Oracle Graph. Oracle AI Vector Search complements the graph model with semantic retrieval so the chatbot can find relevant guidance even when users do not match the source wording. Oracle APEX provides the low-code application layer that exposes the combined graph-and-vector retrieval experience to city employees. This architecture gives support teams a governed RAG-style system that uses context and relationships, not only keywords, to answer internal HR and IT questions and reduce support workload.
Internal HR and IT documents are analyzed by OCI Enterprise AI, which extracts entities and relationships into Oracle Autonomous AI Database and Oracle Graph; Oracle AI Vector Search indexes the same content semantically; Oracle APEX exposes a chatbot that combines graph and vector retrieval to answer employee questions from the centralized knowledge base.
By combining graph and vector queries, the chatbot surfaces related insights that basic RAG can miss.
Oracle APEX ... can answer approximately 85% of support questions.
The solution also uses Oracle AI Vector Search to run semantic similarity searches that can find information by meaning even when keywords don’t match.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Google Cloud Sample Archived · Oct 9, 2026 The State University of New York (SUNY) Now, researchers can skip the technical complexity, giving them more time to focus on what they do best—finding solutions to the world’s biggest problems. 200+ researchers launching more than 80 projects in the first year AI Hypercomputer · Compute Engine · Gemini platform Google Cloud Sample Archived · Oct 2, 2026 Hawaii DCCA The Hawaii Department of Commerce and Consumer Affairs (DCCA) serves a vital mission: protecting residents while ensuring a level playing field that promotes fair business competition across the state. Reclaiming Time: Streamlined operations and reduced manual work across 52 state boards and commissions by… Cloud Storage · Contact Center AI Platform · Customer Experience Agent Studio 
AWS Sample Sep 29, 2026 National Library Board As the National Archives of Singapore prepared to release newly declassified records from The Albatross File, the National Library Board needed to broaden public access to historically important archives without… 100K visitors generating 90,000+ questions through the ChatBook in 5 months AWS · Amazon Bedrock · Anthropic foundation models 
