Random Cut Forest Algorithm
The Random Cut Forest Algorithm (RCF) is an unsupervised algorithm which is used to identify anomalies in data. An anomaly is a data point that differs significantly from the bulk of the data. The Random Cut Forest Algorithm provides a score for each data point. A low score indicates the datapoint is similar to the bulk of the data. A high score indicates an anomaly.
Attributes
| Problem attribute | Description |
| Data types and format | Tabular |
| Learning paradigm or domain | Unsupervised Learning |
| Problem type | Anomaly detection |
| Use case examples | Detect abnormal behavior in an application |
Training
Data for training the Random Cut Forest Algorithm can be in either CSV or x-recordio-protobuf format.
Model artifacts and inference
| Description | Artifacts |
| Learning paradigm | Unsupervised Learning |
| Request format | CSV, JSON, x-recordio-protobuf |
| Result | JSON, x-recordio-protobuf |
Processing environment
CPU instances are used for both Training and Inferencing.
AWS Partner Webinar: Random Cut Forest on Amazon SageMaker
This is a 45.04 minutes video by Chris Burns from AWS.
Detect Anomalies in Your Data with Amazon SageMaker (Level 300)
This is a 29.13 minutes video by Will Badr from AWS.
Credits
Cut down trees photo by Olia Gozha on Unsplash
