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

Generated on: 2025-09-17 13:58:10

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.557
Average intent capture accuracy: 0.560
Average citation accuracy: 0.146
Average document quality score: 4.287
Overall average score: 1.144

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.557 ± 0.202
    Median: 0.563
    Range: 0.199 - 1.000

  intent_capture_accuracy:
    Mean: 0.560 ± 0.171
    Median: 0.600
    Range: 0.200 - 1.000

  citation_accuracy:
    Mean: 0.146 ± 0.199
    Median: 0.000
    Range: 0.000 - 0.571

  document_quality_score:
    Mean: 4.287 ± 0.372
    Median: 4.170
    Range: 4.000 - 5.000

Correlations:
  profile_intent_correlation: -0.499
  intent_quality_correlation: 0.241
  citation_quality_correlation: 0.037
  profile_quality_correlation: -0.101

PERFORMANCE BY DOCUMENT TYPE
------------------------------
email: 1.158 (n=15)
status_report: 1.159 (n=17)
faq: 1.086 (n=8)

PERFORMANCE BY USER ROLE
------------------------------
Project Manager: 1.159 (n=21)
UX Designer: 1.069 (n=7)
Applied Scientist: 1.191 (n=9)
Business Analyst: 1.127 (n=1)
Software Engineer: 1.048 (n=2)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 59.62

Most cited messages:
  Msg_3457: 26 citations
  Msg_2433: 24 citations
  Msg_4169: 23 citations
  Msg_3987: 22 citations
  Msg_2854: 22 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 4.175 ± 0.380
