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

Generated on: 2025-09-17 16:53:14

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.561
Average intent capture accuracy: 0.480
Average citation accuracy: 0.099
Average document quality score: 4.487
Overall average score: 1.143

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.561 ± 0.213
    Median: 0.552
    Range: 0.138 - 1.000

  intent_capture_accuracy:
    Mean: 0.480 ± 0.186
    Median: 0.400
    Range: 0.200 - 0.800

  citation_accuracy:
    Mean: 0.099 ± 0.169
    Median: 0.000
    Range: 0.000 - 0.636

  document_quality_score:
    Mean: 4.487 ± 0.817
    Median: 4.700
    Range: 0.000 - 5.000

Correlations:
  profile_intent_correlation: 0.072
  intent_quality_correlation: -0.123
  citation_quality_correlation: 0.210
  profile_quality_correlation: -0.161

PERFORMANCE BY DOCUMENT TYPE
------------------------------
status_report: 1.224 (n=21)
email: 1.113 (n=13)
faq: 0.927 (n=6)

PERFORMANCE BY USER ROLE
------------------------------
Project Manager: 1.163 (n=23)
Maintenance Engineer: 1.286 (n=5)
Supply Chain Manager: 0.893 (n=3)
Data Analyst: 1.065 (n=1)
Production Manager: 1.153 (n=4)
Quality Engineer: 1.046 (n=4)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 24.07

Most cited messages:
  Msg_748: 75 citations
  Msg_429: 69 citations
  Msg_4344: 34 citations
  Msg_1465: 34 citations
  Msg_1105: 31 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 4.179 ± 0.812
