Concepts#

General concepts#

dataset#

A dataset is a tabular data where:

  • each record must have a timestamp field

  • each column name is called a field.

  • zero or more fields associated with categorical values, called dimension

  • one or more fields associated with quantitative variables, called measure.

record#

Each row in a dataset is called a record.

outlier#

dip, spike

measure#

A measure is a quantitative field in a dataset

dimension#
interval#
metric#

A metric is a combination of a measure value and a dimension value, aggregated within an interval.

Outlier explanation concepts#

predicate#

A predicate is a statement that describes a property or relationship that can hold true or false for a given set of variables. It consists of three parts: a field, an operator, and a value.

  1. The field refers to the type of information that the predicate is describing. For example, a predicate might describe the size of an object, the color of an object, or the relationship between two objects.

  2. The operator is a symbol or word that describes the type of relationship or comparison that is being made between the field and the value. Common operators include equals (=), greater than (>), less than (<), and not equal to (!=).

  3. The value is a specific quantity or quality that the field is being compared to using the operator. For example, if the field is “size” and the operator is “greater than,” the value might be “10.”

Here’s an example of a predicate in logic that uses these three elements: “Size > 10”

This predicate states that the size of something is greater than 10. The field is “size,” the operator is “greater than,” and the value is “10.” This predicate would be true for any object that has a size greater than 10, and false for any object that has a size less than or equal to 10.

conjunction of predicates#

A conjunction of predicates is a group of predicate that are connected by the logical “and” operator. n this case, the “and” operator indicates that all of the predicates in the conjunction must be true for the overall conjunction to be true. or example let’s assume that we have two predicates:

. size > 10 . Client Name == Blackrock

he conjunction of predicates of the two predicates above is True if and only if both the size is strictly greater than 10 and the Client Name is Blackrock.

compound predicate#
decision rule#

decision rule is a group of compound predicates that are used to divide the ensemble into mutually exclusive smaller groups based on the characteristics or features of the data.