Anomaly Detection Configuration
Example Input
[
{
"timestamp": "2023-01-01",
"temperature": 20.5,
"humidity": 45
},
{
"timestamp": "2023-01-02",
"temperature": 21.2,
"humidity": 47
},
{
"timestamp": "2023-01-03",
"temperature": 35.8,
"humidity": 42
}
]Detection Methods:
- Z-Score: Identifies values beyond a number of standard deviations from the mean
- IQR: Uses quartiles to identify values far from the central range
- Isolation Forest: Machine learning approach that isolates anomalies
- Moving Average: Compares values to local averages within a sliding window