feat(scoring): integrated conflict detection and info density metrics v2.71.0

src/retrieval/scoring.ts

@@ -51,7 +51,11 @@ const SCORING_CONFIG = {
     ] as [string, string[]][],
     DENSITY_THRESHOLD: 0.15, // 발췌문 추출 시 최소 키워드 밀도
     TITLE_MULTIPLIER: 3.0,   // 제목 일치 가중치
-    GLOBAL_CACHE_LIMIT: 2000
+    GLOBAL_CACHE_LIMIT: 2000,
+    CONFLICT_INDICATORS: new Set([
+        '반대', '충돌', '오류', '논란', '반박', '차이', '대조',
+        'conflict', 'contradict', 'dispute', 'controversy', 'error', 'mismatch', 'vs'
+    ])
 };
 
 // ─── Global Search State & Cache ───
@@ -168,6 +172,8 @@ export interface ScoredDocument {
     titleBoost: number;
     recencyBoost: number;
     matchedTerms: string[];
+    conflictDetected: boolean;
+    informationDensity: number;
 }
 
 /**
@@ -208,6 +214,9 @@ export function scoreTfIdf(
         let score = 0;
         const matchedTerms: string[] = [];
 
+        // Conflict Detection: 문서 내 상충 지표 확인
+        const conflictDetected = docTokens.some(t => SCORING_CONFIG.CONFLICT_INDICATORS.has(t));
+
         for (const term of expandedQuery) {
             const tf = termFrequency(term, docTokens);
             const idf = idfCache.get(term) || 1;
@@ -217,12 +226,15 @@ export function scoreTfIdf(
                 matchedTerms.push(term);
             }
 
-            // Title match bonus (3x)
-            const titleMultiplier = titleTokens.has(term) ? 3.0 : 1.0;
+            // Title match bonus
+            const titleMultiplier = titleTokens.has(term) ? SCORING_CONFIG.TITLE_MULTIPLIER : 1.0;
             score += tfidf * titleMultiplier;
         }
 
-        // Recency boost: documents modified recently get a boost
+        // Information Density: 쿼리 관련 토큰의 밀도 측정
+        const informationDensity = docTokens.length > 0 ? matchedTerms.length / docTokens.length : 0;
+
+        // Recency boost
         let recencyBoost = 0;
         if (doc.lastModified) {
             const daysAgo = (now - doc.lastModified) / (1000 * 60 * 60 * 24);
@@ -239,7 +251,9 @@ export function scoreTfIdf(
             score: score + recencyBoost + titleBoost,
             titleBoost,
             recencyBoost,
-            matchedTerms: [...new Set(matchedTerms)]
+            matchedTerms: [...new Set(matchedTerms)],
+            conflictDetected,
+            informationDensity
         };
     });
 }