DOCUMENT GENERATION BENCHMARK - DETAILED ANALYSIS REPORT
============================================================

Generated on: 2025-09-17 16:56:22

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.401
Average intent capture accuracy: 0.465
Average citation accuracy: 0.150
Average document quality score: 4.682
Overall average score: 1.175

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.401 ± 0.156
    Median: 0.409
    Range: 0.158 - 0.787

  intent_capture_accuracy:
    Mean: 0.465 ± 0.205
    Median: 0.400
    Range: 0.200 - 1.000

  citation_accuracy:
    Mean: 0.150 ± 0.215
    Median: 0.000
    Range: 0.000 - 0.750

  document_quality_score:
    Mean: 4.682 ± 0.406
    Median: 4.750
    Range: 3.000 - 5.000

Correlations:
  profile_intent_correlation: -0.153
  intent_quality_correlation: 0.078
  citation_quality_correlation: 0.121
  profile_quality_correlation: -0.155

PERFORMANCE BY DOCUMENT TYPE
------------------------------
email: 1.175 (n=15)
status_report: 1.192 (n=19)
faq: 1.117 (n=6)

PERFORMANCE BY USER ROLE
------------------------------
Project Manager: 1.188 (n=31)
Business Analyst: 1.152 (n=5)
IT Systems Lead: 1.101 (n=4)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 22.65

Most cited messages:
  Msg_4372: 52 citations
  Msg_3630: 30 citations
  Msg_3990: 28 citations
  Msg_2772: 22 citations
  Msg_108: 21 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 4.350 ± 0.823
