Skip to main content

Knowledge Graph 25 Classes (Mitsukoshi)

Published 2026-05-14Updated 2026-06-304 min read

Neptune (~550K edges). Reflects VIP + foreign members + 700 tenant brands + duty-free characteristics.

1. 25 Classes​

GroupClasses
Customer / MemberCustomer · Membership (skm-member) · Foreigner (passport, nationality, FX) · Persona · Segment
Product / CatalogProduct · Brand (700+ tenants: LV/Hermes/Chanel/...) · Boutique (in-store counter) · Category · Bundle
Transaction / BehaviorOrderTransaction (online) · POSTransaction (store counter) · TaxRefundTransaction (duty-free) · CartEvent · ReviewRating
Channel / CampaignStore (19 stores) · Campaign (Anniversary Sale / SS / FW) · Promotion · Touchpoint (SMS / KakaoTalk / Xiaohongshu) · Coupon
Operations / ExternalSocialSignal (Xiaohongshu, Dcard, Instagram) · WeatherSignal · EconomicSignal (FX) · TourismSignal (Taiwan Tourism Bureau) · Compliance

2. Mitsukoshi-Specific Classes​

2.1 Foreigner (Foreign Tourist)​

Attributes
foreigner_id (masked) · passport nationality (JP/HK/SG/MY/...) · entry date · TaxRefund applied · FX rate applied

2.2 Boutique (In-store Counter)​

Attributes
boutique_id · brand_id · store_id · floor (1F/2F/...) · area · lease terms

2.3 TaxRefundTransaction (Duty-free Transaction)​

Attributes
txn_id · foreigner_id · txn_amount (KRW / TWD / JPY) · refund_amount · refund_method

2.4 TourismSignal (Tourism Signal)​

Attributes
date · foreign arrivals (by nationality) · average length of stay · average spend

3. Key Relationships​

Edge estimates:

  • Customer × POSTransaction (~200K)
  • Boutique × Brand × Store (~5K)
  • Foreigner × TaxRefundTransaction (~80K)
  • Brand × Product (~100K)
  • External signals (~50K)
  • Other (~120K)

→ ~550K edges

4. openCypher Examples​

4.1 S9-M — Duty-free Recommendation for Japanese Women in Their 30s​

MATCH (f:Foreigner {nationality: 'JP'})-[:CLASSIFIED_AS]->(p:Persona {age_band: '30s', gender: 'F'})
MATCH (f)-[:MADE]->(t:TaxRefundTransaction)-[:CONTAINS]->(:Product)-[:OF_BRAND]->(b:Brand)
RETURN b.name, count(t) AS popularity ORDER BY popularity DESC LIMIT 10

4.2 S10-M — Luxury SOV on the 1F of Xinyi Store​

MATCH (s:Store {name: '台北信義'})<-[:IN_STORE]-(bq:Boutique {floor: '1F'})
      -[:OF_BRAND]->(b:Brand)
MATCH (bq)<-[:AT_BOUTIQUE]-(t:POSTransaction)
WHERE t.paid_at > datetime() - duration('P30D')
RETURN b.name, sum(t.total_amount) AS gmv
ORDER BY gmv DESC

4.3 S11-M — Pre-sale Recommendation for Black VIP Members​

MATCH (c:Customer)-[:HAS_MEMBERSHIP]->(m:Membership {grade: 'Black'})
MATCH (c)-[:MADE]->(:POSTransaction)-[:CONTAINS]->(p:Product)
      -[:OF_BRAND]->(b:Brand)
WITH c, b, count(*) AS frequency
ORDER BY frequency DESC LIMIT 5
RETURN c.customer_id, collect(b.name)[..5] AS preferred_brands

5. cohort_tag​

ValueMeaning
realPII-masked members (N=2K Customer + 500 Foreigner)
synth49.5K synthetic
externalSocial, FX, tourism, weather

6. OpenSearch Indices​

IndexAnalyzer
idx_productSmartcn (Traditional Chinese) + Standard
idx_customerSmartcn
idx_review (Xiaohongshu / Dcard external)Smartcn + Kuromoji (Japanese)
idx_brandStandard (brand name + metadata)