Price Sentiment

Anomaly detection has become imperative in business

What is an anomaly?

Determining what's an anomaly is not an easy task neither for humans or machines.

Why companies need anomaly detection?

Types of anomaly detectors

Univariate

can have any number of sources, and as they ingest data they proceed to
normalize it into [events](#events) \(see next section\). This sets the stage

for easy and consistent processing of your data. Examples of sources include file, syslog, socket, and stdin.

View all sources

Multivarite

A "transform" are machine learning models that can learn how to tranforms raw unstructured data from

View all transforms

Contextual Anomalies

A "destination" is a destination for events. Each sink's interacting with. For example, the [socket sink][docs.sinks.÷] will design and transmission method is dictated by the downstream service it is stream individual events, while the aws_s3 sink will buffer and flush data.

View all sinks

Real-Time Anomaly

All items (both logs and metrics) passing through Vector are known as an "event", which is explained in detail in the data model section.

View data model

Out of the box

A "pipeline" is the end result of connecting sources, transforms, and sinks. You can see a full example of a pipeline in the configuration section.

View configuration

Ou tof the box

A "pipeline" is the end result of connecting sources, transforms, and sinks. You can see a full example of a pipeline in the configuration section.

View configuration