Package

weka.filters.unsupervised.attribute

Synopsis

An filter that assumes instances form time-series data and replaces attribute values in the current instance with the difference between the current value and the equivalent attribute attribute value of some previous (or future) instance. For instances where the time-shifted value is unknown either the instance may be dropped, or missing values used. Skips the class attribute if it is set.

Options

The table below describes the options available for TimeSeriesDelta.

Option	Description
attributeIndices	Specify range of attributes to act on. This is a comma separated list of attribute indices, with "first" and "last" valid values. Specify an inclusive range with "-". E.g: "first-3,5,6-10,last".
fillWithMissing	For instances at the beginning or end of the dataset where the translated values are not known, use missing values (default is to remove those instances)
instanceRange	The number of instances forward/backward to merge values between. A negative number indicates taking values from a past instance.
invertSelection	Invert matching sense. ie calculate for all non-specified columns.

Capabilities

The table below describes the capabilites of TimeSeriesDelta.

Capability	Supported
Class	Nominal class, Date class, Unary class, Empty nominal class, Binary class, String class, Numeric class, No class, Relational class, Missing class values
Attributes	Numeric attributes, Nominal attributes, Relational attributes, Binary attributes, Missing values, Date attributes, String attributes, Empty nominal attributes, Unary attributes
Min # of instances	0