US 2017/0249562 A1Published application

Supervised method for classifying seasonal patterns

AssigneeOracle International Corporation

InventorsDustin Garvey, Uri Shaft, Lik Wong

Life of this patent

Estimating the term…

Loading patent…

Abstract

Techniques are described for classifying seasonal patterns in a time series. In an embodiment, a set of time series data is decomposed to generate a noise signal and a dense signal. Based on the noise signal, a first classification is generated for a plurality of seasonal instances within the set of time series data, where each respective instance of the plurality of instances corresponds to a respective sub-period within the season and the first classification associates a first set of one or more instances from the plurality of instances with a particular class of seasonal pattern. Based on the dense signal, a second classification is generated that associates a second set of one or more instances with the particular class. Based on the first classification and the second classification, a third classification is generated, where the third classification associates a third set of one or more instances with the particular class.

Claims20

Dependent claims

A method comprising:

generating, for a set of time series data, a noise signal and a dense signal;
wherein the noise signal includes sparse features from the set of time series data;
wherein the dense signal includes dense features from the set of time series data;
generating, based on the noise signal, a first classification for a plurality of instances of a season within the set of time series data;
wherein each respective instance of the plurality of instances corresponds to a respective sub-period within the season;
wherein the first classification for the plurality of instances of the season associates a first set of one or more instances from the plurality of instances with a particular class of seasonal pattern;
generating, based on the dense signal, a second classification for the plurality of instances of the season;
wherein the second classification for the plurality of instances of the season associates a second set of one or more instances from the plurality of instances with the particular class of seasonal pattern;
generating and storing, in volatile or non-volatile storage, a third classification for the plurality of instances of the season based on the first classification and the second classification;
wherein the third classification for the plurality of instances of the season associates a third set of one or more instances from the plurality of instances with the particular class of seasonal pattern.

7 more elements

One or more non-transitory computer-readable media storing instructions, wherein the instructions include:

instructions for generating, for a set of time series data, a noise signal and a dense signal;
wherein the noise signal includes sparse features from the set of time series data;
wherein the dense signal includes dense features from the set of time series data;
instructions for generating, based on the noise signal, a first classification for a plurality of instances of a season within the set of time series data;
wherein each respective instance of the plurality of instances corresponds to a respective sub-period within the season;
wherein the first classification for the plurality of instances of the season associates a first set of one or more instances from the plurality of instances with a particular class of seasonal pattern;
instructions for generating, based on the dense signal, a second classification for the plurality of instances of the season;
wherein the second classification for the plurality of instances of the season associates a second set of one or more instances from the plurality of instances with the particular class of seasonal pattern;
instructions for generating and storing, in volatile or non-volatile storage, a third classification for the plurality of instances of the season based on the first classification and the second classification;
wherein the third classification for the plurality of instances of the season associates a third set of one or more instances from the plurality of instances with the particular class of seasonal pattern.

7 more elements

Dependencies are inferred from the claim text.