ML Market Map — Korea Equity Clusters for 2026-08-31

A daily unsupervised machine-learning read of the Korea Equity market: 1,155 stocks grouped into 7 consensus clusters (KMeans + Gaussian-mixture + hierarchical, over robust-scaled PCA features) for 2026-08-31. Descriptive, not predictive — there is no buy or sell signal. Research, not investment advice.

Markets: US Equity · China A-Share · Japan Equity · Hong Kong Equity · Taiwan Equity · Korea Equity · UK Equity · India Equity

Today's clusters

• high margin · high cash yield · cheap (high E/P) — 465 names; mostly Industrials; drivers: profit margin (+0.62), val cfop z (+0.58), cfo growth (+0.56)
• high leverage · deep pullbacks (21d) · expensive (low E/P) — 348 names; mostly Industrials; drivers: leverage debt to mcap (+0.85), cfo growth (−0.57), pa mae 21d (−0.52)
• strong 1y momentum · wide 1y upside range · favorable 1y edge — 127 names; mostly Technology; drivers: ret 252d (+2.41), pa mfe 52w (+1.95), pa edge 52w (+1.60)
• low margin · high turnover · attention spent — 111 names; mostly Healthcare; drivers: profit margin (−3.33), turnover to mcap (+1.47), pa sir recovered (+1.37)
• deep pullbacks (21d) · attention spent · expensive (low E/P) — 60 names; mostly Consumer Cyclical; drivers: pa mae 21d (−2.80), pa sir recovered (+2.76), vol of vol 63 (+2.75)
• rising (3m) · high volatility (3m) · high volatility (1y) — 29 names; mostly Industrials; drivers: ret 63d (+4.53), vol of vol 63 (+4.07), ret 126d (+3.94)
• attention spent · high turnover · wide upside excursions (21d) — 15 names; mostly Industrials; drivers: pa sir recovered (+4.28), turnover to mcap (+3.58), ret 20d (+2.18)

Machine-readable data (free, read-only JSON)

The full map, per-ticker cluster assignments with confidence and anomaly scores, and PCA structure are published as open JSON for automated and AI-analyst consumption:

• /api/ml/latest?market=kr — clusters, names, sector mix, representative tickers, top features
• /api/ml/tickers?market=kr — per-ticker cluster, confidence, anomaly score, day-over-day migration, tenure
• /api/ml/pca?market=kr — PCA components, explained variance, feature loadings

Descriptive market-structure research only. Unsupervised clustering finds structure, not direction; a tight cluster or an anomaly is a starting point for research, never a trade signal.