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

Generated on: 2025-09-17 17:59:30

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.595
Average intent capture accuracy: 0.455
Average citation accuracy: 0.209
Average document quality score: 4.254
Overall average score: 1.150

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.595 ± 0.242
    Median: 0.542
    Range: 0.158 - 1.000

  intent_capture_accuracy:
    Mean: 0.455 ± 0.169
    Median: 0.400
    Range: 0.200 - 0.800

  citation_accuracy:
    Mean: 0.209 ± 0.247
    Median: 0.062
    Range: 0.000 - 0.767

  document_quality_score:
    Mean: 4.254 ± 0.300
    Median: 4.300
    Range: 3.700 - 4.700

Correlations:
  profile_intent_correlation: -0.005
  intent_quality_correlation: 0.237
  citation_quality_correlation: 0.191
  profile_quality_correlation: 0.168

PERFORMANCE BY DOCUMENT TYPE
------------------------------
status_report: 1.237 (n=19)
email: 1.073 (n=14)
faq: 1.070 (n=7)

PERFORMANCE BY USER ROLE
------------------------------
Supply Chain Manager: 1.261 (n=6)
Project Manager: 1.104 (n=15)
Maintenance Engineer: 1.324 (n=5)
Reliability Engineer: 1.188 (n=2)
Product Manager: 1.230 (n=1)
Production Manager: 1.169 (n=5)
Quality Engineer: 0.995 (n=5)
Business Analyst: 0.834 (n=1)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 85.95

Most cited messages:
  Msg_437: 63 citations
  Msg_2519: 43 citations
  Msg_2851: 33 citations
  Msg_4104: 32 citations
  Msg_3844: 27 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 3.875 ± 0.399
