US 2019/0095805 A1 · Ensembled Decision Systems Using Feature Hashing Models

Abstract

Systems and methods are disclosed to build and execute a decision system based on multiple machine learned decision models. In embodiments, the decision system performs a hashing technique to reduce relevant features of the input data into a feature vector for each decision model. The feature vector reduces the dimensionality of the feature universe of the input data, and its use allows the decision models to be trained and executed using less computing resources. In embodiments, the decision system implements an ensembled decision model that makes decisions based on a combination function that combines the decision results of the individual models in the ensemble. The decision models employ different hashing techniques to hash the input features differently, so that errors caused by the feature hashing of individual models are reduced in the aggregate.

Claims21

Dependent claims

A system, comprising:

one or more processors with associated memory implementing a decision system comprising:
a plurality of decision models trained using a machine learning technique, each configured to, in parallel:
perform a respective hashing technique to reduce features of an input data to a feature vector;
and
generate a decision result based on the feature vector;
and
a results combiner configured to combine the decision results of the plurality of decision models to generate a decision, wherein an error rate of the decision system is less than individual error rates of the decision models.

4 more elements

A non-transitory, computer-readable storage medium storing program instructions that when executed on one or more computers cause the one or more computers to:

generate a plurality of respective decision results from a plurality of decision models in parallel, wherein each decision model is trained using a machine learning technique and configured to:
perform a respective hashing technique to reduce features of an input data to a feature vector;
and
generate the decision result based on the feature vector;
and
combine the decision results of the plurality of decision models to generate a decision, wherein an error rate of the decision is less than individual error rates of the decision results.

3 more elements

Dependencies are inferred from the claim text.

Ensembled Decision Systems Using Feature Hashing Models

Life of this patent

Ensembled Decision Systems Using Feature Hashing Models

Life of this patent

Abstract

Claims21