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

Generated on: 2025-09-17 17:07:42

BASIC STATISTICS
------------------------------
Total queries processed: 40
Average user profile accuracy: 0.397
Average intent capture accuracy: 0.490
Average citation accuracy: 0.135
Average document quality score: 4.432
Overall average score: 1.117

ADVANCED METRICS
------------------------------
Score Statistics:
  user_profile_accuracy:
    Mean: 0.397 ± 0.175
    Median: 0.379
    Range: 0.109 - 0.735

  intent_capture_accuracy:
    Mean: 0.490 ± 0.202
    Median: 0.500
    Range: 0.200 - 1.000

  citation_accuracy:
    Mean: 0.135 ± 0.183
    Median: 0.000
    Range: 0.000 - 0.500

  document_quality_score:
    Mean: 4.433 ± 0.482
    Median: 4.600
    Range: 3.300 - 5.000

Correlations:
  profile_intent_correlation: -0.146
  intent_quality_correlation: -0.236
  citation_quality_correlation: 0.285
  profile_quality_correlation: 0.202

PERFORMANCE BY DOCUMENT TYPE
------------------------------
status_report: 1.164 (n=17)
email: 1.077 (n=16)
faq: 1.094 (n=7)

PERFORMANCE BY USER ROLE
------------------------------
Data Analyst: 1.084 (n=3)
Business Analyst: 1.180 (n=7)
Project Manager: 1.088 (n=18)
IT Systems Lead: 1.135 (n=4)
Risk Analyst: 1.164 (n=4)
Product Manager: 1.093 (n=4)

CITATION ANALYSIS
------------------------------
Documents with citations: 40/40
Average citations per document: 20.40

Most cited messages:
  Msg_473: 74 citations
  Msg_727: 28 citations
  Msg_3670: 27 citations
  Msg_860: 17 citations
  Msg_566: 16 citations

QUALITY DIMENSIONS ANALYSIS
------------------------------
citation_quality: 3.950 ± 0.835
