← Back to home

🇻🇳 Vietnamese Address Parser — Free-Text → Structured JSON

Parse Vietnamese free-text addresses into structured JSON: street, ward, district, province. Supports 63 provinces + 30 HCMC districts + 30 Hanoi districts + aliases (Sài Gòn/HCM, Huế/TT-Huế). Includes confidence score.

63 provincesHCM + HN districtsAliasesJSON output
Samples:

Why use this tool

🎯
Parse messy addresses

User types '227 NVC, Q5, HCMC' → JSON {street, district, province}. No hand-rolled regex.

🇻🇳
VN-standard aliases

Sài Gòn → Hồ Chí Minh, Nha Trang → Khánh Hòa, Đà Lạt → Lâm Đồng, Huế → Thừa Thiên Huế, Hạ Long → Quảng Ninh. Normalizes to official names.

📊
Confidence score

High (3+ fields filled), Medium (2 fields), Low (1 field). Backends can reject low-confidence parses and ask users to confirm.

How to use

  1. 1Paste a free-text address.
  2. 2Tool parses in real time → 4 fields + confidence.
  3. 3Copy the JSON for code use.

Parsing Vietnamese addresses — a hard problem

Vietnamese ecommerce / logistics / CRM frequently need to parse free-form addresses into structure to (a) compute ship fees per province, (b) match warehouses, (c) check serviceability, (d) cluster customers by geography. Users type all kinds of formats: '227 Nguyễn Văn Cừ, Quận 5, TP.HCM', 'Số 1 Đại Cồ Việt P. Bách Khoa Q. Hai Bà Trưng Hà Nội', '12 Hà Huy Tập BMT Đắk Lắk' — all must parse correctly.

Tool uses rule-based matching: 63 provinces + aliases + HCMC/Hanoi district lists. Districts of other provinces fall into 'street' (incomplete). Phase 2 will add the full 712-district + 11k-ward DB from the General Statistics Office.

vs Google / VietMap geocoding APIs: this tool is free, rate-limit-free, key-free, privacy-safe. Trade-off: Medium-Low confidence needs user confirmation, no lat/lng output.

  • 63 Vietnamese provinces
  • 30 HCMC districts + 30 Hanoi districts
  • Aliases Sài Gòn / Nha Trang / Đà Lạt / Huế / Hạ Long
  • High/medium/low confidence
  • JSON output
  • 100% client-side, no API key

FAQ

Can it parse ward/commune details?

Phase 1 only parses wards via the keyword 'phường/xã/thị trấn' + a single token. Phase 2 will embed the full 11k administrative unit DB.

Lat/lng output?

No — the tool doesn't call geocoding APIs (privacy + cost). For lat/lng, use Google Geocoding or Mapbox separately.

Why does 'Sài Gòn' map to 'Hồ Chí Minh'?

The official administrative name is 'Hồ Chí Minh' (renamed 1976). The tool normalizes aliases to official names for backend matching.