Statistical Model
Mathematical description of data over time
4 Keys of Anomaly Detection
Feature selectionLoading Existing DataPeriod DetectionEmpirical Sorting
Feature selection
Creating detection rules off of your data points
Period Detection
Choosing what frequency of time is meaningful to the data.
Empirical scoring
Defining which automatically detected anomalies are important.

Bucket Spans
Parameter range that divides data into batches for processing (usually time)
Analysis function
Function that is applied to the bucket span, count, sum, metric etc.
Machine Learning Job
Running of the analysis function(s) over the determined bucket spans
Detector
Multiple analysis functions with a shared bucket span.
Event/Data Feed
Pushes data into a job.May be a query.Can be range based or live.Can be run on pre-aggregated data.

Influencer
Attribute that has an influence on the data, something that has contributed to the anomaly
Anomaly Score
Combination of individual item scoring and bucket scoring
Individual item scoring
How anomalous an event is to a baseline.Based on past behavior
Bucket scoring
Comparing an anomaly to other readings in the bucket. Aggregate score across all detectors for the jobOnly one score per bucket.
Partition fields
Fields with a low enough cardinality to run a machine learning job over each distinct value.

Advanced job parameter that duplicates the job over values of the field and performs analysis in parallel.

Individual Anomaly Detection
Comparison of a behavior to it's own historical behavior
Population Anomaly Detection
Comparison of a behavior to the behavior of other members of a population.
Over Fields
Advanced job parameter used to perform population anomaly detection.
By fields
Advanced job parameter used to perform individual anomaly detection. Jobs are performed in serial, then categorized BY field.
By field anomaly score
Uses total anomaly ratio for all by field values
Partition field anomaly score
Uses individual anomaly ratio for all partition field values