Real-time prediction and explanation of sequences of abnormal events

Patent metadata

Field	Value
Publication number	US-10824647-B2
Application number	US-201715816774-A
Country	US
Kind code	B2
Filing date	Nov 17, 2017
Priority date	Nov 17, 2017
Publication date	Nov 3, 2020
Grant date	Nov 3, 2020

Abstract

Implementations are directed to providing a semantic sequence based on a sequence associated with the event, and context data provided from a knowledge graph, receiving a set of cases associated with at least one entity included in the event, the set of cases including previous instances of events, and actions performed to resolve the previous instances of events, defining a sub-set of cases from the set of cases based on the context data, for each case of the sub-set of cases, determining a similarity score, each similarity score representing a degree of similarity between the event, and a case of the sub-set of cases, determining an explanation based on features of cases in the sub-set of cases based on the context data, and providing one or more actions based on actions of cases in the sub-set of cases.

First claim

What is claimed is:

1. A computer-implemented method for automated provision of an explanation for an event, and one or more actions to resolve the event, the method being executed by one or more processors and comprising:
processing, by an event detector, input data to detect an occurrence of an event
in response to detecting the occurrence of the event, providing a semantic sequence based on a sequence of entities associated with the event, and context data provided from a knowledge graph, the knowledge graph comprising domain-specific context data that is specific to a domain associated with the event;
receiving a set of cases associated with at least one entity included in the event, cases in the set of cases comprising previous instances of events that have occurred and that are provided based on historical data, and actions that had been performed to resolve the previous instances of events;
defining a sub-set of cases from the set of cases based on the context data;
for each case of the sub-set of cases, determining a similarity score, each similarity score representing a degree of similarity between the event, and a case of the sub-set of cases;
generating an explanation based on features of cases in the sub-set of cases based on the context data, the explanation comprising a concatenation of a set of features including a first feature of a first case in the sub-set of cases and a second feature of a second case in the sub-set of cases, the first feature and the second feature selected for inclusion in the explanation based on respective feature scores each exceeding a threshold feature score, the explanation explaining reasons for the occurrence of the event;
providing one or more actions that can be performed to resolve the event, the one or more actions being provided based on actions of cases in the sub-set of cases; and
outputting the explanation and the one or more actions as output data.

2. The method of claim 1 , further comprising receiving input data from one or more domain-specific data sources, at least a portion of the input data being received in real-time.

3. The method of claim 1 , wherein the similarity scores are determined based on a respective plurality of temporal similarity scores, each temporal similarity score representing a degree of similarity between features within a temporal window of the event, and a respective case.

4. The method of claim 1 , wherein the explanation is determined from features of cases in the sub-set of cases based on respective feature scores, and subsumption of two or more features.

5. The method of claim 1 , wherein the one or more actions are provided based on subsumption of two or more actions provided in the set of cases.

6. The method of claim 1 , wherein the explanation and the one or more actions are transmitted to a computer-implemented system.

7. The method of claim 1 , wherein the event and the knowledge graph correspond to an airline flight domain.

8. The method of claim 7 , wherein the event comprises one of a flight delay, and a flight cancellation.

9. The method of claim 1 , wherein, for each case in the set of cases, determining a temporal sequence of the events.

10. The method of claim 9 , further comprising providing a knowledge graph based on the temporal sequence of the events.

11. A non-transitory computer-readable storage medium coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for automated provision of an explanation for an event, and one or more actions to resolve the event, the operations comprising:
processing, by an event detector, input data to detect an occurrence of an event in response to detecting the occurrence of the event, providing a semantic sequence based on a sequence of entities associated with the event, and context data provided from a knowledge graph, the knowledge graph comprising domain-specific context data that is specific to a domain associated with the event;
receiving a set of cases associated with at least one entity included in the event, cases in the set of cases comprising previous instances of events that have occurred and that are provided based on historical data, and actions that had been performed to resolve the previous instances of events;
defining a sub-set of cases from the set of cases based on the context data;
for each case of the sub-set of cases, determining a similarity score, each similarity score representing a degree of similarity between the event, and a case of the sub-set of cases;
generating an explanation based on features of cases in the sub-set of cases based on the context data, the explanation comprising a concatenation of a set of features including a first feature of a first case in the sub-set of cases and a second feature of a second case in the sub-set of cases, the first feature and the second feature selected for inclusion in the explanation based on respective feature scores each exceeding a threshold feature score, the explanation explaining reasons for the occurrence of the event;
providing one or more actions that can be performed to resolve the event, the one or more actions being provided based on actions of cases in the sub-set of cases; and
outputting the explanation and the one or more actions as output data.

12. The non-transitory computer-readable storage medium of claim 11 , wherein operations further comprise receiving input data from one or more domain-specific data sources, at least a portion of the input data being received in real-time.

13. The non-transitory computer-readable storage medium of claim 11 , wherein the similarity scores are determined based on a respective plurality of temporal similarity scores, each temporal similarity score representing a degree of similarity between features within a temporal window of the event, and a respective case.

14. The non-transitory computer-readable storage medium of claim 11 , wherein the explanation is determined from features of cases in the sub-set of cases based on respective feature scores, and subsumption of two or more features.

15. The non-transitory computer-readable storage medium of claim 11 , wherein the one or more actions are provided based on subsumption of two or more actions provided in the set of cases.

16. The non-transitory computer-readable storage medium of claim 11 , wherein the explanation and the one or more actions are transmitted to a computer-implemented system.

17. The non-transitory computer-readable storage medium of claim 11 , wherein the event and the knowledge graph correspond to an airline flight domain.

18. The non-transitory computer-readable storage medium of claim 17 , wherein the event comprises one of a flight delay, and a flight cancellation.

19. The non-transitory computer-readable storage medium of claim 11 , wherein, for each case in the set of cases, determining a temporal sequence of the events.

20. The non-transitory computer-readable storage medium of claim 19 , wherein operations further comprise providing a knowledge graph based on the temporal sequence of the events.

21. A system, comprising:
one or more processors; and
a non-transitory computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for automated provision of an explanation for an event, a

Assignees

• Accenture Global Solutions Ltd

Inventors

• Lecue Freddy
• Jiménez Alejandro Cabello

Classifications

• G06Q10/04 Primary

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title

• G06F16/24575

  using context · CPC title

• G06F16/285 Primary

  Clustering or classification · CPC title

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title