AI search understanding for maps
Tasco · Posted 2 months ago
Automotive and mobilitySearch & Knowledge Retrieval
- Budget
- To be determined
- Timeline
- Not specified
- Closed on
- 31 Jul 2026, 16:59 ICT
Problem statement
Vietnamese map search has typos, abbreviations, missing accents, incomplete addresses, ambiguous names, and mixed-language queries.
Expected outcomes and success metrics
- Turn noisy search input into structured search-ready intents.
- Improve Vietnamese query handling.
- Support map search APIs.
Data availability and readiness
Search Query Dataset covering Vietnamese typos, abbreviations, natural language, incomplete addresses, and ambiguous queries.
POI Dataset with names, categories, brands, and addresses.
Address Dataset with Vietnamese addresses and address variations.
Abbreviation Dictionary with common abbreviations used in Vietnamese map search.
---
*Track: Mobility | Sponsor: Tasco | Slug: mobility:maps-search-understanding | Source: Agentic AI Build Week 2026 — Production snapshot 2026-08-11*
Integration, deployment and infrastructure
Query Input: user search query.
Normalization Layer: lowercase, accent normalization.
Spell Correction Layer: typo correction.
Abbreviation Layer: abbreviation expansion.
Address Parser: Vietnamese address parsing.
Query Understanding Layer: intent and entity extraction.
Search API: structured query output.
Relevant AI technologies
Search & Knowledge Retrieval