Predicting and classifying network activity events

Patent metadata

Field	Value
Publication number	US-9727882-B1
Application number	US-201313924046-A
Country	US
Kind code	B1
Filing date	Jun 21, 2013
Priority date	Jun 21, 2013
Publication date	Aug 8, 2017
Grant date	Aug 8, 2017

Abstract

Disclosed are various embodiments for predicting and classifying events that create a sudden and substantial increase in network traffic activity. To begin, a sales event may be predicted based on the results following the search for one or more prediction factors that occur during a predefined period of time. Based either on the individual results of each evaluation or a combination of results of the search of two or more of the prediction factors, a sales event may be predicted. Additionally, upon detection of a sudden and substantial increase in network traffic activity, one or more classification factors may be evaluated to determine whether the cause of the increase is due to a sales event or other type of alternative activity event.

First claim

Therefore, the following is claimed:

1. A non-transitory computer-readable medium embodying a program executable in at least one computing device, wherein, when executed, the program causes the at least one computing device to at least:
identify a plurality of event prediction factors probative in predicting whether a future sales event will occur during a predefined future period of time, the future sales event being related to an offering of an item for sale in an electronic marketplace that potentially alters a network activity of an electronic commerce system associated with the electronic marketplace during the predefined future period of time;
determine an event probability value associated with the future sales event based at least in part upon a sum of weighted values associated with the plurality of event prediction factors, the event probability value reflecting a probability of an occurrence of the future sales event during the predefined future period of time;
store in a data store a notation indicating a prediction that the future sales event will occur during the predefined future period of time in response to the event probability value meeting or exceeding a predefined event threshold value;
monitor an amount of the network activity of the electronic commerce system;
detect a network activity event during the predefined future period of time in the electronic marketplace based at least in part on an increase in the amount of the network activity of the electronic commerce system detected while monitoring the amount of the network activity;
determine that the network activity event is the future sales event based at least in part upon the network activity event occurring during the predefined future period of time and the notation indicating the prediction that the future sales event will occur during the predefined future period of time; and
adjust the electronic commerce system to support the increase in the network activity by adjusting one or more network requirements associated with the electronic commerce system in response to determining that the network activity event is the future sales event, the one or more network requirements comprising at least one of a network capacity, a router activity, or a cache load.

2. The non-transitory computer-readable medium of claim 1 , wherein the plurality of event prediction factors comprise at least one of a sales promotion, a change in an item price, an item messaging event, an availability of item inventory, a social media interest, or an increase in the item inventory.

3. The non-transitory computer-readable medium of claim 1 , wherein determining whether the network activity event is the future sales event is further based at least in part upon at least one of an increase of current order sales associated with the electronic marketplace, a predefined number of interested users interacting with the electronic marketplace, or whether the network activity event is determined to be associated with at least one sale referrer.

4. A system, comprising:
at least one computing device; and
at least one application executable in the at least one computing device, the at least one application further causing the at least one computing device to at least:
identify an event prediction factor probative in predicting whether a future sales event will occur during a predefined future period of time, wherein a network activity associated with an electronic commerce system is potentially altered due to the future sales event;
determine an event probability value based at least in part upon the event prediction factor, the event probability value reflecting a likelihood that the future sales event will occur during the predefined future period of time;
store in a data store a notation indicating a prediction that the future sales event will occur during the predefined future period of time when the event probability value reaches a predefined event threshold value;
monitor an amount of the network activity associated with the electronic commerce system;
determine that an increase in the amount of the network activity associated with the electronic commerce system corresponds to the future sales event based at least in part upon whether the increase of the network activity occurs during the predefined future period of time and the notation indicating the prediction that the future sales event will occur during the predefined future period of time; and
adjust the electronic commerce system to support the increase in the amount of the network activity by adjusting one or more network requirements associated with the electronic commerce system, the one or more network requirements comprising at least one of a network capacity, a router activity, or a cache load.

5. The system of claim 4 , wherein the future sales event is related to an offering of an item for sale in an electronic marketplace that results in the increase in the network activity.

6. The system of claim 4 , wherein the event prediction factor is one of a sales promotion, a change in an item price, an item messaging event, an availability of item inventory, a social media interest, or an inventory increase in the item inventory.

7. The system of claim 4 , wherein the event prediction factor is one of a plurality of event prediction factors, and the event probability value is determined by a sum of weighted values associated with the plurality of event prediction factors, and the sum of weighted values is based at least in part upon a combination of the plurality of event prediction factors.

8. The system of claim 4 , wherein the event probability value is further based at least in part upon a popularity of an item associated with the future sales event.

9. The system of claim 4 , wherein the event probability value is increased by a prediction weight corresponding to the event prediction factor, and the prediction weight is based at least in part upon at least one characteristic associated with at least one prior sales event.

10. A method, comprising:
identifying, via at least one of one or more computing devices, a plurality of event prediction factors that are probative in predicting whether a future sales event will occur during a predefined future period of time, wherein a network activity of an electronic commerce system supporting an electronic marketplace is potentially altered due to the future sales event;
determining, via at least one of the one or more computing devices, an event probability value that represents a likelihood of the future sales event occurring during the predefined future period of time, the event probability value being based at least in part upon a sum of weighted values associated with at least one of the plurality of event prediction factors;
monitoring, via at least one of the one or more computing devices, an amount of the network activity of the electronic commerce system;
detecting, via at least one of the one or more computing devices, an increase in the amount of the network activity of the electronic commerce system; and
adjusting, via at least one of the one or more computing devices, the electronic commerce system to support the increase in the amount of the network activity by adjusting at least a subset of a plurality of network requirements associated with the electronic marketplace during at least the predefined future period of time in response to the event probability value being greater than an event threshold value, and the plurality of network requirements being associated with at least one of a network capacity, a router activity, or a cache load.

11. The method of claim 10 , wherein the future sales event is related to an offering

Assignees

• Amazon Tech Inc

Inventors

• Huntwork Andrew Stuart
• Davis Melissa Elaine
• Jewell Gavin Richard

Classifications

• H04L43/0876

  Network utilisation, e.g. volume of load or congestion level · CPC title

• G06Q30/0202 Primary

  Market predictions or forecasting for commercial activities · CPC title

• G06F9/50

  Allocation of resources, e.g. of the central processing unit [CPU] · CPC title