Applied artificial intelligence technology for runtime computation of story outlines to support natural language generation (NLG)
Artificial intelligence (AI) technology can be used in combination with composable communication goal statements and an ontology to facilitate a user's ability to quickly structure story outlines in a manner usable by an NLG narrative generation system without any need for the user to directly author computer code. This AI technology permits NLG systems to determine the appropriate content for inclusion in a narrative story about a data set in a manner that will satisfy a desired communication goal. The determined content can be arranged in a computed story outline that is created at runtime, and NLG can be performed on the computed story outline to generate the narrative story.
1. A method for artificial intelligence-based natural language generation (NLG), the method comprising:
a processor selecting and parameterizing a set of narrative analytics based on a plurality of communication goal statements;
a processor executing the selected and parameterized narrative analytics set with reference to a structured data set to determine content for inclusion in a narrative story about the structured data;
a processor creating a computed story outline in response to the executing, the computed story outline comprising the determined content in an arrangement that defines a narrative structure for the narrative story, wherein the determined content comprises information linked to (1) a plurality of elements of the structured data set and (2)a plurality of ontological objects within an ontology, wherein the ontology defines a knowledge base about the structured data set for narrative story generation; and
a processor performing NLG on the computed story outline with reference to the ontology to generate the narrative story such that the narrative story uses the ontological objects to express the determined content.
2. The method of claim 1 wherein the communication goal statements comprise a plurality of communication goal statement parameters, wherein the communication goal statement parameters are linked to a plurality of ontological objects within the ontology, and wherein the selecting and parameterizing step comprises a processor parameterizing the set of narrative analytics based on the communication goal statement parameters.
3. The method of claim 2 wherein the communication goal statements comprise a plurality of communication goal statement operators, and wherein the selecting and parameterizing step comprises a processor selecting the set of narrative analytics based on the communication goal statement operators.
4. The method of claim 2 wherein the selecting step further comprises a processor selecting the set of narrative analytics based on the communication goal statement operators and the communication goal statement parameters.
5. The method of claim 2 further comprising:
a processor composing at least one of the communication goal statements in response to user input.
6. The method of claim 5 wherein the composing step comprises a processor composing at least one of the communication goal statements in response to natural language user input.
7. The method of claim 2 wherein the communication goal statements have an associated order, and wherein the creating step comprises a processor determining the arrangement for the determined content in the computed story outline based on the associated order for the communication goal statements.
8. The method of claim 1 wherein the ontology comprises a plurality of expressions associated with the ontological objects, and wherein the performing step comprises a processor selecting expressions for referring to the linked ontological objects in the narrative story from among the expressions associated with the linked ontological objects in the ontology.
9. The method of claim 8 wherein the ontological objects comprise (1) a plurality of entity types, and (2) a plurality of attributes for a plurality of the entity types.
10. The method of claim 9 wherein the ontological objects further comprise at least one characterization for at least one of the entity types.
11. The method of claim 9 wherein the ontological objects further comprise at least one relationship between at least two of the entity types.
12. The method of claim 1 wherein the communication goal statements include a communication goal statement for presenting a value of an attribute of an entity.
13. The method of claim 12 wherein the communication goal statements further include a communication goal statement for comparing a value of the attribute of the entity to another value of another attribute of the entity.
14. The method of claim 1 wherein the communication goal statements include a communication goal statement for presenting a characterization of an entity.
15. The method of claim 1 wherein the communication goal statements include a communication goal statement for comparing a value of an attribute of an entity to another value of an attribute of the entity.
16. The method of claim 1 wherein the communication goal statements include a communication goal statement for calling out an entity.
17. An artificial intelligence (AI) natural language generation (NLG) computer system, the system comprising:
a memory configured to store an ontology and a plurality of narrative analytics, wherein the ontology defines a knowledge base about a structured data set for narrative story generation;
a processor for cooperation with the memory, the processor configured to:
select and parameterize a set of the narrative analytics based on a plurality of communication goal statements;
execute the selected and parameterized narrative analytics set with reference to the structured data set to determine content for inclusion in a narrative story about the structured data;
create a computed story outline in response to the execution of the selected and parameterized narrative analytics set, the computed story outline comprising the determined content in an arrangement that defines a narrative structure for the narrative story, wherein the determined content comprises information linked (1) a plurality of elements of the structured data set and (2) a plurality of ontological objects within the ontology; and
perform NLG on the computed story outline with reference to the ontology to generate the narrative story such that the narrative story uses the ontological objects to express the determined content.
18. A computer program product for an artificial intelligence (AI) natural language generation (NLG) computer system, the computer program product comprising:
a plurality of instructions resident on a non-transitory computer-readable storage medium, wherein the instructions are executable by a processor to a cause a processor to:
select and parameterize a set of narrative analytics based on a plurality of communication goal statements;
execute the selected and parameterized narrative analytics set with reference to a structured data set to determine content for inclusion in a narrative story about the structured data;
create a computed story outline in response to the executing, the computed story outline comprising the determined content in an arrangement that defines a narrative structure for the narrative story, wherein the determined content comprises information linked (1) a plurality of elements of the structured data set and (2) a plurality of ontological objects within an ontology, wherein the ontology defines a knowledge base about the structured data set for narrative story generation; and
perform NLG on the computed story outline with reference to the ontology to generate the narrative story such that the narrative story uses the ontological objects to express the determined content.
19. The system of claim 17 wherein the communication goal statements comprise a plurality of communication goal statement parameters, wherein the communication goal statement parameters are linked to a plurality of ontological objects within the ontology, and wherein the processor is further configured to parameterize the set of narrative analytics based on the communication goal statement parameters.
20. The system of claim 17 wherein the ontology comprises a plurality of expressions associated with the ontological objects, and wherein the processor is further configured to select expressions for referring to the linked ontological objects in the narrative story from among the expressions associated with the linked ontological objects in the ontology.