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

Generated on: 2025-09-17 13:48:19

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.600
Average intent capture accuracy: 0.490
Average citation accuracy: 0.107
Average document quality score: 4.196
Overall average score: 1.109

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.600 ± 0.241
    Median: 0.571
    Range: 0.193 - 1.000

  intent_capture_accuracy:
    Mean: 0.490 ± 0.157
    Median: 0.400
    Range: 0.200 - 0.800

  citation_accuracy:
    Mean: 0.107 ± 0.183
    Median: 0.000
    Range: 0.000 - 0.750

  document_quality_score:
    Mean: 4.196 ± 0.287
    Median: 4.000
    Range: 4.000 - 5.000

Correlations:
  profile_intent_correlation: 0.127
  intent_quality_correlation: 0.302
  citation_quality_correlation: -0.117
  profile_quality_correlation: -0.075

PERFORMANCE BY DOCUMENT TYPE
------------------------------
status_report: 1.155 (n=19)
email: 1.034 (n=14)
faq: 1.135 (n=7)

PERFORMANCE BY USER ROLE
------------------------------
Supply Chain Manager: 1.251 (n=6)
Project Manager: 1.065 (n=16)
Maintenance Engineer: 1.195 (n=7)
Data Analyst: 1.023 (n=1)
Production Manager: 1.068 (n=5)
Quality Engineer: 1.014 (n=5)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 55.00

Most cited messages:
  Msg_3871: 22 citations
  Msg_3882: 22 citations
  Msg_3366: 21 citations
  Msg_4351: 20 citations
  Msg_2677: 20 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 4.075 ± 0.263
