US 8,805,767 B1Granted

Machine learning memory management and distributed rule evaluation

AssigneeAmazon Technologies, Inc.

InventorsJue Wang, Madhu M. Kurup, Jared S. Lundell, Thomas Robert Park

Life of this patent

Estimating the term…

Loading patent…

Abstract

Aspects of the present disclosure relate to management of evaluated rule data sets. Specifically, a unreduced evaluated rule data set may contain a number of items to be compared or analyzed according to a number of rules, and may also contain the results of such analysis. An illustrative reduced evaluated data set can include the results of evaluated rules. When utilized in conjunction with an item data set and a rule data set, the information contained within the unreduced evaluated rule data set may be maintained. The reduce memory requirements of the reduced evaluated rule data set may facilitate storage of the reduced evaluated rule data set in faster to access memory, or may facilitate distributed computation of the reduced evaluated rule data set.

Claims25

Dependent claims

A computer-implemented method for distributive creation of a reduced evaluated rule data set, the computer-implemented method comprising:

determining a set of item combination identifiers, wherein each item combination identifier corresponds to a combination of items to be evaluated against a set of rules for creation of a reduced evaluated rule data set;
determining a set of rule identifiers, wherein each rule identifier corresponds to a rule used in evaluating items for creation of the reduced evaluated rule data set, and wherein each rule comprises a set of rule components;
for each pair of (1) an item combination identifier from the set of item combination identifiers and (2) a rule identifier from the set of rule identifiers:
transmitting the pair of item combination identifier and rule identifier to at least one rule evaluation computing component of a set of rule evaluation computing components;
receiving a result of an evaluation based at least in part on the item combination identifier and the rule identifier;
aggregating the received results to create a reduced evaluated rule data set, wherein each result within the reduced evaluated rule data set is indexed according to a corresponding pair of item combination identifier and rule identifier;
and
storing the reduced evaluated rule data set.

5 more elements

A system for creation of a reduced evaluated rule data set, the system comprising:

one or more computing devices configured to:
determine a set of item combination identifiers, wherein each item combination identifier corresponds to a combination of items to be evaluated against a set of rules;
determine a set of rule identifiers, wherein each rule identifier corresponds to a rule comprising a set of rule components;
for each pair of (1) an item combination identifier from the set of item combination identifiers and (2) a rule identifier from the set of rule identifiers, evaluate the combination of items corresponding to the item combination identifier based at least in part on the rule corresponding to the rule identifier;
aggregate each results of each evaluation to create a reduced evaluated rule data set, wherein each result within the reduced evaluated rule data set is indexed according to a corresponding pair of item combination identifier and rule identifier;
and
cause the storage of the reduced evaluated rule data set.

4 more elements

A non-transitory computer readable medium having at least one computer executable component for creation of a reduced evaluated data set, the at least one computer executable component comprising:

an evaluation management component configured to:
determine a set of item combination identifiers, wherein each item combination identifier corresponds to a combination of items;
determine a set of condition identifiers, wherein each condition identifier corresponds to a condition;
for each pair of (1) an item combination identifier from the set of item combination identifiers and (2) a condition identifier from the set of condition identifiers, evaluate the combination of items corresponding to the item combination identifier based at least in part on the condition corresponding to the condition identifier;
create a reduced evaluated data set comprising a result of each evaluation, wherein each result within the reduced evaluated data set is indexed according to a corresponding pair of item combination identifier and condition identifier;
and
store the reduced evaluated data set.

4 more elements

Dependencies are inferred from the claim text.