Automatic prediction of an event using data

Patent metadata

Field	Value
Publication number	US-10839296-B2
Application number	US-201615365362-A
Country	US
Kind code	B2
Filing date	Nov 30, 2016
Priority date	Nov 30, 2016
Publication date	Nov 17, 2020
Grant date	Nov 17, 2020

Abstract

A device may receive data associated with an event. The device may identify a context of the event based on receiving the data. The device may identify a similar event based on performing a comparison of the context of the event and a context of the similar event. The device may determine a set of pre-events associated with the event based on identifying a pre-event that occurred before the similar event. The set of pre-events may include at least one pre-event similar to the pre-event that occurred before the similar event. The device may determine a set of post-events associated with the event based on determining the set of pre-events and identifying a post-event that occurred after the similar event. The set of post-events may include at least one post-event similar to the post-event. The device may perform an action based on the set of post-events.

First claim

What is claimed is:

1. A device, comprising:
a memory; and
one or more processors to:
receive data associated with a first event;
identify a first context of the first event based on the data;
identify a plurality of second events based on the first context and a second context of the plurality of second events,
the first context being semantically similar to the second context;

use a homophily technique to determine a score for the first event and the plurality of second events based on identifying the plurality of second events,
the score indicating a semantic similarity between the first event and a second event, of the plurality of second events, based on a knowledge graph,
the knowledge graph to store data from the first event and the plurality of second events in a plurality of nodes,
a node, of the plurality of nodes, including a geographic proximity of the first event and the plurality of second events,
the score being based on the plurality of nodes, and

the homophily technique being represented by:

S
⁡

(

E
i

,

E
j

,
KG

)

=

(

total
⁢

⁢
quantity
⁢

⁢
of
⁢

⁢
connected
⁢

⁢
nodes

total
⁢

⁢
quantity
⁢

⁢
of
⁢

⁢
nodes
⁢

⁢
in
⁢

⁢
knowledge
⁢

⁢
graph

)

where S represents the score and indicates a similarity between the first event and each second event based on the knowledge graph, where Ei represents event i, which is the first event, Ej represents event j, which is each second event of the plurality of second events, KG represents a relationship between data in the knowledge graph for events i and j, total quantity of connected nodes represents a total quantity of nodes in the knowledge graph that data related to events i and j have in common, and events i and j are different events;

determine an order for the plurality of second events based on the score indicating a semantic similarity between the first event and each second event of the plurality of second events;
determine a set of pre-events associated with the first event based on a plurality of pre-events associated with the plurality of second events and based on the order for the plurality of second events,
one or more pre-events of the set of pre-events being similar to the plurality of pre-events;

identify a plurality of post-events associated with the plurality of second events based on identifying the plurality of second events in the knowledge graph;
identify another plurality of post-events associated with a plurality of third events based on identifying the plurality of third events in the knowledge graph,
a third context of the plurality of third events being similar to the first context,
the first context, the second context, and the third context each including one or more of:
information associated with a date of occurrence,
information associated with a time of occurrence,
information associated with a location of occurrence, or
a device identifier of a device from which data associated with the first event was sent;

determine a set of post-events associated with the first event based on the set of pre-events and based on the plurality of post-events associated with the plurality of second events and the other plurality of post-events,
the set of post-events including one or more post-events predicted to occur after the first event,
the one or more post-events being similar to the plurality of post-events; and

perform an action related to the first event based on determining the set of post-events.

2. The device of claim 1 , where the one or more processors are further to:
determine a timing of steps of the plurality of second events; and
where the one or more processors, when determining the set of pre-events, are to:
determine the set of pre-events based on the timing of the steps of the plurality of second events.

3. The device of claim 1 , where the one or more processors are to:
use a canonical correlation technique to identify the plurality of second events based on receiving the data;
apply a filter to the plurality of second events based on identifying the plurality of second events,
the filter including a temporal filter or a geographic filter; and

where the one or more processors, when identifying the plurality of second events, are to:
identify the plurality of second events based on applying the filter to the plurality of second events.

4. The device of claim 1 , where the one or more processors are further to:
identify a second action associated with the plurality of second events based on identifying the plurality of second events;
determine a set of actions associated with the first event based on identifying the second action; and
where the one or more processors, when performing the action, are to:
perform the action based on determining the set of actions,
the action being included in the set of actions.

5. The device of claim 1 , where the one or more processors are further to:
determine a severity level of the set of post-events; and
where the one or more processors, when performing the action, are to:
perform the action based on the severity level of the set of post-events.

6. A method, comprising:
receiving, by a device, data associated with an event;
identifying, by the device, a context of the event based on receiving the data;
identifying, by the device, a similar event based on performing a comparison of the context of the event and a context of the similar event,
each context including one or more of:
information associated with a date of occurrence,
information associated with a time of occurrence,
inf

Assignees

• Accenture Global Solutions Ltd

Inventors

• Verdejo Dominique
• Lecue Freddy
• Zaman Md Faisal

Classifications

• G06V10/84

  using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks · CPC title

• G06N5/022 Primary

  Knowledge engineering; Knowledge acquisition · CPC title

• G06N5/02 Primary

  Knowledge representation; Symbolic representation · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F18/29

  Graphical models, e.g. Bayesian networks · CPC title