Method and apparatus for motion description

• US 9,990,360 B2
• Filed: 12/27/2012
• Issued: 06/05/2018
• Est. Priority Date: 12/27/2012
• Status: Active Grant

• Current Assignee: Arria Data2text Limited
• Original Assignee: Arria Data2text Limited
• Inventors: Sripada, Gowri Somayajulu
• Primary Examiner(s): SPOONER, LAMONT M
• Application Number:
  US14/650,777
• Publication Number:
  US 20150347400A1
• Time in Patent Office: 1,986 Days
• Field of Search: 704 1, 704 9, 704 10, 707706-708
• CPC Class Codes: G06F 40/56 Natural language generation

Abstract

A method, apparatus, and computer program product for describing motion. The method may include receiving a set of eventualities (114). The set of eventualities (114) may describe at least one of a domain event and a domain state. The at least one of the domain event and the domain state may be derived from a set of spatio-temporal data (102) and the set of eventualities (114) may be associated with a particular region and a particular time period. The method may include organizing the set of eventualities to generate a document plan. The method may further include generating, using a processor, a linguistic representation of the set of eventualities using the document plan.

Claims

1. An apparatus that is configured to transform an input data stream comprising spatio-temporal data that is expressed at least in part in a non-linguistic format into a format that can be expressed at least in part via a linguistic representation in a textual output, the apparatus comprising:

a memory coupled to at least one processor; and

the at least one processor, configured to;

receive a set of eventualities, the set of eventualities describing at least one of a domain event and a domain state, the at least one of the domain event and the domain state derived from a set of spatio-temporal data and the set of eventualities associated with a particular region and a particular time period;

organize the set of eventualities according to a domain model;

wherein organizing the set of eventualities comprises

determining an importance score for one or more of the set of eventualities using the domain model that comprises a set of importance rules for one or more of the set of eventualities, wherein the importance rules provide an importance score based on an externally specified importance value for an eventuality type, a number of spatial points in the eventuality, and a time period of the eventuality;

organizing the set of eventualities based on the importance scores; and

at least one of filtering out one or more eventualities, partitioning one or more of the set of eventualities into a portion of the particular region, and ordering the set of eventualities into a particular order;

generate a document plan, wherein the document plan is generated based on the organized set of eventualities;

instantiate the document plan with one or more messages that describe each eventuality of the organized set of eventualities; and

generate a linguistic representation of the one or more messages using the document plan, wherein the linguistic representation of the one or more messages is displayable via a user interface.

2. The apparatus of claim 1, wherein the particular region is a geographic region.

3. The apparatus of claim 1, wherein the processor is further configured to organize the set of eventualities based on the importance by placing a most important eventuality first in the document plan.

4. The apparatus of claim 1, wherein the most important eventuality is placed first in the document plan in response to determining that a difference in an importance score between the most important eventuality'"'"'s importance score and a next most important eventuality'"'"'s importance score is greater than a threshold importance score value.

5. The apparatus of claim 1, wherein the domain model is associated with a domain of the set of eventualities.

6. The apparatus of claim 5, wherein the domain of the set of eventualities is at least one of weather data, traffic data, medical data, scientific data, and computer network data.

7. The apparatus of claim 1, wherein organizing the set of eventualities further comprises ordering the set of eventualities based on a start time of one or more of the set of eventualities.

8. The apparatus of claim 1, wherein organizing the set of eventualities comprises partitioning the set of eventualities into one or more portions of the region, and the set of eventualities are organized in the document plan based on a respective portion of the region into which each eventuality was partitioned.

9. The apparatus of claim 8, wherein the processor is further configured to order the partitions into a particular order to improve coherence of the document plan.

10. The apparatus of claim 1, wherein the processor is further configured to organize the set of eventualities by:

attempting to separate the set of eventualities into one or more portions of the region;

determining that it is not possible to separate the set of eventualities into one or more portions of the region; and

in response to determining that it is not possible to separate the set of eventualities into one or more portions of the region, organizing the document plan as a single partition.

11. The apparatus of claim 1, wherein the processor is further configured to generate the linguistic representation by:

conducting document planning, microplanning, and realization using the one or more messages and the document plan to result in an output text.

12. A non-transitory computer readable storage medium that is configured to transform an input data stream comprising spatio-temporal data that is expressed at least in part in a non-linguistic format into a format that can be expressed at least in part via a linguistic representation in a textual output, the non-transitory computer readable storage medium comprising instructions, that, when executed by a processor, configure the processor to:

receive a set of eventualities, the set of eventualities describing at least one of a domain event and a domain state, the at least one of the domain event and the domain state derived from a set of spatio-temporal data and the set of eventualities associated with a particular region and a particular time period;

organize the set of eventualities according to a domain model;

wherein organizing the set of eventualities comprises

determining an importance score for one or more of the set of eventualities using the domain model that comprises a set of importance rules for one or more of the set of eventualities, wherein the importance rules provide an importance score based on an externally specified importance value for an eventuality type, a number of spatial points in the eventuality, and a time period of the eventuality;

organizing the set of eventualities based on the importance scores; and

at least one of filtering out one or more eventualities, partitioning one or more of the set of eventualities into a portion of the particular region, and ordering the set of eventualities into a particular order;

generate a document plan, wherein the document plan is generated based on the organized set of eventualities;

instantiate the document plan with one or more messages that describe each eventuality of the organized set of eventualities; and

generate a linguistic representation of the one or more messages using the document plan, wherein the linguistic representation of the one or more messages is displayable via a user interface.

13. The non-transitory computer readable storage medium of claim 12, wherein the particular region is a geographic region.

14. The non-transitory computer readable storage medium of claim 12, wherein the non-transitory computer readable storage medium further comprises instructions to configure the processor to organize the set of eventualities based on the importance by placing a most important eventuality first in the document plan.

15. The non-transitory computer readable storage medium of claim 12, wherein the most important eventuality is placed first in the document plan in response to determining that a difference in an importance score between the most important eventuality'"'"'s importance score and a next most important eventuality'"'"'s importance score is greater than a threshold importance score value.

16. The non-transitory computer readable storage medium of claim 12, wherein the domain model is associated with a domain of the set of eventualities.

17. The non-transitory computer readable storage medium of claim 16, wherein the domain of the set of eventualities is at least one of weather data, traffic data, medical data, scientific data, and computer network data.

18. The non-transitory computer readable storage medium of claim 12, wherein organizing the set of eventualities further comprises ordering the set of eventualities based on a start time of one or more of the set of eventualities.

19. The non-transitory computer readable storage medium of claim 12, wherein organizing the set of eventualities comprises partitioning the set of eventualities into one or more portions of the region, and the set of eventualities are organized in the document plan based on a respective portion of the region into which each eventuality was partitioned.

20. The non-transitory computer readable storage medium of claim 19, further comprising instructions to order the partitions into a particular order to improve coherence of the document plan.

21. The non-transitory computer readable storage medium of claim 12, further comprising instructions to configure the processor to organize the set of eventualities by:

attempting to separate the set of eventualities into one or more portions of the region;

determining that it is not possible to separate the set of eventualities into one or more portions of the region; and

in response to determining that it is not possible to separate the set of eventualities into one or more portions of the region, organizing the document plan as a single partition.

22. The non-transitory computer readable storage medium of claim 12, further comprising program instructions to generate the linguistic representation by:

conducting document planning, microplanning, and realization using the one or more messages and the document plan to result in an output text.