Advanced CO₂ & Sensor Anomaly Detection (C2D)
Purpose of the Algorithm
This algorithm performs second‑order analytics to detect anomalies that emerge only through statistical comparison, correlation, and distribution analysis.
It is intended as an advanced or premium Compute‑to‑Data component.
What the Algorithm Does
Extreme value detection
- Identifies CO₂ spikes using percentile thresholds (P01 / P99)
- Flags rare or extreme environmental events
Anomaly aggregation
- Counts and classifies anomaly frequency
- Supports anomaly density metrics
Correlation insights
- Enables correlation between battery behavior and CO₂ anomalies
- Supports drift and instability analysis
Dataset health indicators
- Produces compact anomaly KPIs
- Enables “data quality score” extensions
Value in a Compute‑to‑Data Workflow
- High insight density with minimal compute cost
- No raw values exposed
- Ideal for premium analytics tiers
Industry Value
Supports:
- Environmental risk detection
- Dataset certification
- Advanced marketplace products
- Trust and compliance layers