US 2017/0372222 A1Published application

Technologies for detection of minority events

AssigneeIntel Corporation

InventorsVarvara Kollia, Ramune Nagisetty

Life of this patent

Estimating the term…

Loading patent…

Abstract

Technologies for detecting minority events are disclosed. By performing a guided hierarchical classification algorithm with a decision tree structure and grouping the minority class(es) in with some of the majority classes, large majority classes may be separated from a minority class without requiring good detection of the minority events by themselves. The decision tree structure may be used only for the purpose of identifying if the data sample in question is a member of a minority class. If it is determined that it is not, a primary classification algorithm may be used. With this approach, the guided hierarchical classification algorithm need not perform as well as the primary classification algorithm for the majority events, but may provide improved detection for minority events.

Claims25

Dependent claims

A compute device for classifying minority events, the compute comprising:

a data acquisition module to acquire training data comprising a plurality of training data samples, the plurality of training data samples comprising a plurality of labeled training data samples, each labeled training data sample belonging to a class of a plurality of classes, the plurality of classes comprising one or more minority classes and a plurality of majority classes;
a minority classification module to:
train a guided hierarchical classification algorithm comprising a decision tree structure and a plurality of nodes of the decision tree structure, each node of the plurality of nodes comprising one or more classes of the plurality of classes, wherein at least two nodes of the plurality of nodes comprise at least one minority class of the one or more minority classes and at least one majority class of the plurality of majority classes,
wherein to train the guided hierarchical classification algorithm comprises to distribute, for each of the at least two nodes, the classes in the corresponding node to two or more corresponding child nodes based on the number of training data samples in each of the classes in the corresponding node.

1 more element

A method for classifying minority events with a compute device, the method comprising:

acquiring, by the compute device, training data comprising a plurality of training data samples, the plurality of training data samples comprising a plurality of labeled training data samples, each labeled training data sample belonging to a class of a plurality of classes, the plurality of classes comprising one or more minority classes and a plurality of majority classes;
training, by the compute device, a guided hierarchical classification algorithm comprising a decision tree structure and a plurality of nodes of the decision tree structure, each node of the plurality of nodes comprising one or more classes of the plurality of classes, wherein at least two nodes of the plurality of nodes comprise at least one minority class of the one or more minority classes and at least one majority class of the plurality of majority classes,
wherein training the guided hierarchical classification algorithm comprises distributing, for each of the at least two nodes, the classes in the corresponding node to two or more corresponding child nodes based on the number of training data samples in each of the classes in the corresponding node.

One or more computer readable media comprising a plurality of instructions stored thereon that, when executed, cause a compute device to:

acquire training data comprising a plurality of training data samples, the plurality of training data samples comprising a plurality of labeled training data samples, each labeled training data sample belonging to a class of a plurality of classes, the plurality of classes comprising one or more minority classes and a plurality of majority classes;
and
train a guided hierarchical classification algorithm comprising a decision tree structure and a plurality of nodes of the decision tree structure, each node of the plurality of nodes comprising one or more classes of the plurality of classes, wherein at least two nodes of the plurality of nodes comprise at least one minority class of the one or more minority classes and at least one majority class of the plurality of majority classes,
wherein to train the guided hierarchical classification algorithm comprises to distribute, for each of the at least two nodes, the classes in the corresponding node to two or more corresponding child nodes based on the number of training data samples in each of the classes in the corresponding node.

1 more element

Dependencies are inferred from the claim text.