Sector Classification
Classifies companies into investment sectors based on the WICS (Wise Industry Classification Standard). Uses a 3-tier priority system and provides sector-specific valuation parameters.
Usage
from dartlab.engines.sector import classify, getParams
info = classify("삼성전자", kindIndustry="통신 및 방송 장비 제조업")
info.sector # Sector.IT
info.industryGroup # IndustryGroup.SEMICONDUCTOR
info.confidence # 1.0
info.source # "override"
params = getParams(info)
params.discountRate # 13.0 (%)
params.perMultiple # 15
params.label # "반도체"
classify()
classify(companyName: str, kindIndustry: str = None, mainProducts: str = None) -> SectorInfo
| Parameter | Type | Description |
|---|
companyName | str | Company name |
kindIndustry | str (optional) | KIND industry name (KSIC-based) |
mainProducts | str (optional) | Main products text |
3-Tier Classification Priority
| Tier | Method | Confidence | Target |
|---|
| 1 | Manual override | 0.95~1.0 | ~100 large caps (삼성전자, SK하이닉스, etc.) |
| 2 | Keyword analysis | 0.6~0.9 | Keyword matching from main products text |
| 3 | KSIC mapping | 0.7 | KIND industry name to WICS mapping (200+ items) |
If no match is found at any tier, returns Sector.UNKNOWN (confidence 0.0).
SectorInfo
| Field | Type | Description |
|---|
sector | Sector | WICS major classification (11 sectors) |
industryGroup | IndustryGroup | WICS mid-level classification (47 groups) |
confidence | float | Confidence score (0.0~1.0) |
source | str | Classification basis (“override”, “keyword”, “ksic”, “unknown”) |
Sector (11 Major Classifications)
| Value | Korean | English |
|---|
ENERGY | 에너지 | Energy |
MATERIALS | 소재 | Materials |
INDUSTRIALS | 산업재 | Industrials |
CONSUMER_DISC | 경기관련소비재 | Consumer Discretionary |
CONSUMER_STAPLES | 필수소비재 | Consumer Staples |
HEALTHCARE | 건강관리 | Healthcare |
FINANCIALS | 금융 | Financials |
IT | IT | IT |
COMMUNICATION | 커뮤니케이션서비스 | Communication Services |
UTILITIES | 유틸리티 | Utilities |
REAL_ESTATE | 부동산 | Real Estate |
getParams()
getParams(sectorInfo: SectorInfo) -> SectorParams
Uses mid-level parameters if available; otherwise falls back to major classification parameters.
SectorParams
| Field | Type | Description |
|---|
discountRate | float | Discount rate (%) |
growthRate | float | Growth rate (%) |
perMultiple | float | PER multiple |
pbrMultiple | float | PBR multiple |
evEbitdaMultiple | float | EV/EBITDA multiple |
label | str | Sector label (Korean) |
Sector parameters are benchmark values reflecting industry characteristics. Used in DCF, PER/PBR band analysis, etc.
Examples
from dartlab.engines.sector import classify, getParams, Sector
# Large cap — Immediately classified via manual override
info = classify("SK하이닉스")
print(info) # SectorInfo(IT/반도체, conf=1.00, src=override)
# Mid/small cap — Keyword-based classification
info = classify("에코프로비엠", mainProducts="양극재, 이차전지 소재")
print(info) # SectorInfo(IT/2차전지, conf=0.90, src=keyword)
# KSIC-based classification
info = classify("미래에셋증권", kindIndustry="증권 중개업")
print(info) # SectorInfo(FINANCIALS/증권, conf=0.70, src=ksic)
# Valuation parameters
params = getParams(info)
print(f"PER: {params.perMultiple}, Discount rate: {params.discountRate}%")
