buzzline-06-world

Data Definitions

This document includes:

1. Core Data Fields
2. Aggregated Metrics
3. Account Behavior Metrics
4. Suggested Detection Thresholds

1. Core Data Fields

1.1 Individual API Data Points (and Example Interpretations)

ts (timestamp)

• Type: Timestamp
• Description: UTC timestamp for when this data point was observed (ISO 8601 format)
• Example: 2026-02-01T09:45:00Z
• Usage: Temporal analysis of coordination patterns and event correlation

volume

• Type: Integer
• Description: Total number of posts observed in this time interval (typically 1 minute)
• Range: 50-800+ posts per minute
• Interpretation:
  • 50-200: Normal organic activity
  • 300-800: Potential coordination burst

burst_score

• Type: Float (0-1)
• Description: Burstiness indicator measuring deviation from normal posting patterns
• Range: 0.05-0.95
• Interpretation:
  • 0.05-0.25: Organic, steady posting
  • 0.35-0.60: Moderate coordination
  • 0.75-0.95: High coordination burst

synchrony_index

• Type: Float (0-1)
• Description: Temporal synchronization measure indicating coordinated timing
• Range: 0.10-0.90
• Interpretation:
  • 0.10-0.35: Natural human posting rhythms
  • 0.40-0.65: Some coordination
  • 0.70-0.90: High synchronization (suspicious)

duplication_clusters

• Type: Integer
• Description: Count of near-duplicate content clusters detected
• Range: 1-15 clusters per minute
• Interpretation:
  • 1-4: Normal content diversity
  • 4-8: Moderate recycling
  • 8-15: Mass content duplication (coordination indicator)

topic (content hash)

• Type: String (8 characters)
• Description: Privacy-preserving hash identifier for content cluster
• Format: Lowercase hexadecimal (e.g., a7f3d9e2)
• Purpose: Enables analysis without revealing actual content

2. Aggregated Metrics

2.1 Coordination Summary Fields (Suggested Signals That Might Assist With Analysis)

total_points

• Type: Integer
• Description: Total number of data points (minutes) analyzed for this content hash
• Typical Value: 900 (15 topics × 5 time windows × 12 hours of data)

avg_burst_score

• Type: Float (0-1)
• Description: Average burst score across all time intervals for this content hash
• Interpretation:
  • ~0.15: Organic discourse baseline
  • ~0.60: Coordinated manipulation

max_burst_score

• Type: Float (0-1)
• Description: Highest burst score observed for this content hash
• Purpose: Identifies peak coordination moments

high_burst_count

• Type: Integer
• Description: Number of intervals with burst_score > 0.7 (coordination threshold)
• Interpretation:
  • 0: No coordination events
  • 250-300: Systematic coordination campaign

avg_synchrony_index

• Type: Float (0-1)
• Description: Average temporal synchronization across all intervals
• Purpose: Measures overall coordination consistency

avg_duplication_clusters

• Type: Float
• Description: Average number of content duplication clusters per interval
• Interpretation:
  • 2-4: Natural content sharing
  • 7-8: Coordinated content recycling

suspicion_score

• Type: Float (0-1)
• Description: Composite coordination indicator combining multiple signals
• Formula: (burst_score * 0.4) + (synchrony_index * 0.4) + (duplication_factor * 0.2)
• Interpretation:
  • 0.18-0.19: Organic civic discourse
  • 0.59-0.60: Coordinated manipulation

3. Account Behavior Metrics and Example Interpretations

avg_new_accounts

• Type: Float (0-1)
• Description: Average proportion of posts from accounts 0-7 days old
• Interpretation:
  • ~0.10: Normal community growth
  • ~0.27: New account surge (manipulation indicator)

avg_api_posting

• Type: Float (0-1)
• Description: Average proportion of posts made via API clients
• Interpretation:
  • ~0.10: Mostly manual posting
  • ~0.28: High automation (coordination indicator)

new_account_automation_correlation

• Type: Float (0-1)
• Description: Correlation between new account activity and API client usage
• Purpose: Detects coordinated account creation + automation

4. Example Detection Thresholds

4.1 Example Coordination Indicators (what would you use?)

• High burst events: burst_score > 0.7
• High synchrony events: synchrony_index > 0.7
• High duplication events: duplication_clusters > 8
• Coordinated burst flag: burst_score > 0.7 AND synchrony_index > 0.7
• New account surge: avg_new_accounts > 0.3
• High automation: avg_api_posting > 0.4

4.2 Example Manipulation Score Calculation (what would you use?)

manipulation_score = (burst_score * 0.3) + 
                     (synchrony_index * 0.3) + 
                     (min(duplication_clusters/15, 1.0) * 0.2) +
                     (automation_ratio * 0.1) +
                     (new_account_ratio * 0.1)

This site is open source. Improve this page.