Interactive and conversational data exploration
Example embodiments are disclosed where a narrative generation platform is integrated with a conversational interface such as a Chatbot to support interactive and dynamic narrative generation in response to speech inputs from users. Such a system can be further integrated with a visualization platform to allow the speech input through the conversational interface to interactively and dynamically focus and control visualizations and/or narratives that accompany the visualizations.
1. A method comprising:
a processor interactively generating a narrative story about a data set based on a series of sequentially received natural language inputs from a user, wherein the narrative story comprises a series of unfolding natural language narrative responses to the natural language inputs that form an interactive dialog relating to the data set, wherein each natural language input has a corresponding natural language narrative response and is indicative of a communication goal to be addressed by its corresponding natural language narrative response, wherein the interactively generating step comprises:
in response to each of the natural language inputs, the processor (1) receiving data derived from that natural language input, (2) based on the received data, parameterizing a narrative analytics model relevant to the communication goal indicated by that natural language input, (3) executing the parameterized narrative analytics model with respect to the data set to determine content about the data set for an outline of the corresponding natural language narrative response, and (4) generating the corresponding natural language narrative response based on the outline so that the corresponding natural language narrative response addresses the communication goal indicated by that natural language input and expresses the determined content in natural language.
2. The method of claim 1 wherein the interactively generating step further comprises the processor interfacing with a natural language conversational interface to receive the data derived from the natural language inputs, wherein the natural language conversational interface sequentially receives the series of natural language inputs.
3. The method of claim 2 wherein the data set comprises visualization parameter data.
4. The method of claim 2 wherein the received data comprises a plurality of commands derived from the natural language inputs.
5. The method of claim 4 wherein at least one of the commands comprises a focus task or a drill down task.
6. The method of claim 4 wherein at least one of the commands includes a specification of an entity.
7. The method of claim 6 wherein the entity comprises a specific entity identified in at least one of the natural language inputs.
8. The method of claim 6 wherein the entity comprises a notional entity that is notionally identified in at least one of the natural language inputs.
9. The method of claim 2 further comprising:
the processor (1) deriving information from logic that the processor uses to generate natural language, and (2) providing the derived information to the natural language conversational interface; and
the natural language conversational interface defining a language pattern for identifying a pattern in natural language input based on the provided information.
10. The method of claim 9 wherein the generating step includes the processor determining a natural language expression for referring to an element in at least one of the natural language narrative responses so that that the natural language expression serves as the derived information;
wherein the providing step comprises the processor providing the determined natural language expression to the natural language conversational interface; and
the method further comprising the natural language conversational interface using the provided natural language expression as a language pattern for identifying a reference in at least one of the natural language inputs to the element.
11. The method of claim 10 wherein the element comprises an entity.
12. The method of claim 11 wherein the entity comprises a notional entity.
13. The method of claim 9 wherein the derived information comprises information about a command corresponding to a natural language input so that the identified pattern comprises the corresponding command.
14. The method of claim 2 further comprising:
communicating the natural language narrative responses to the natural language conversational interface for sequential presentation to the user via the natural language conversational interface as part of the interactive dialog.
15. The method of claim 1 wherein at least one of the narrative analytics models and outlines includes a specification of a notional entity;
wherein the executing step includes the processor determining a specific entity to which the notional entity refers based on an analysis of the data set; and
wherein the generating step comprises the processor determining a natural language expression for referring to the specific entity corresponding to the notional entity.
16. The method of claim 15 wherein the specific entity comprises a plurality of specific entities to which the notional entity refers.
17. The method of claim 15 wherein the interactively generating step further comprises the processor interfacing with a natural language conversational interface to receive the data derived from the natural language inputs, wherein the natural language conversational interface sequentially receives the series of natural language inputs, the method further comprising:
the processor providing the determined natural language expression to the natural language conversational interface.
18. The method of claim 17 further comprising:
the natural language conversational interface using the provided natural language expression as a language pattern for identifying a reference in natural language input to the specific entity corresponding to the notional entity.
19. The method of claim 1 wherein at least one of the narrative analytics models and outlines includes a specification of a notional entity;
wherein the generating step includes the processor determining a natural language expression for referring to the notional entity.
20. The method of claim 19 wherein the interactively generating step further comprises the processor interfacing with a natural language conversational interface to receive the data derived from the natural language inputs, wherein the natural language conversational interface sequentially receives the series of natural language inputs, the method further comprising:
the processor providing the determined natural language expression to the natural language conversational interface.
21. The method of claim 20 further comprising:
the natural language conversational interface using the provided natural language expression as a language pattern for identifying a reference in natural language input to the notional entity.
22. The method of claim 1 wherein at least one of the natural language inputs comprises speech input.
23. The method of claim 1 wherein at least one of the natural language inputs comprises text input.
24. The method of claim 1 wherein the processor is part of a narrative generation artificial intelligence platform.
25. The method of claim 1 wherein the processor comprises a plurality of processors.
26. The method of claim 1 wherein the determined content includes a characterization of the data set derived from execution of at least one of the parameterized narrative analytics models.
27. The method of claim 1 wherein the communication goals addressed by the narrative story as part of the interactive dialog include communication goals to track a performance of an entity, evaluate the performance of the entity, and explain the performance of the entity.
28. A method comprising:
configuring a narrative generation artificial intelligence platform with operating parameters from a natural language conversational interface as part of an interactive dialog with a user, wherein the interactive dialog relates to a data set, wherein the interactive dialog includes a series of natural language inputs from the user to the natural language conversational interface, wherein the operating parameters are derived from the natural language inputs, and wherein the natural language inputs are indicative of a plurality of communication goals to be addressed by the interactive dialog; and
the configured narrative generation artificial intelligence platform generating a narrative story about the data set based on the operating parameters, wherein the narrative story is generated as a series of unfolding natural language narrative responses to the natural language inputs by translating the operating parameters into a plurality of parameterized narrative analytics models that are executed with respect to the data set to determine content for outlines of elements to be expressed in the natural language narrative responses, wherein the narrative analytic models and outlines are associated with the communication goals indicated by the natural language inputs, and wherein the natural language narrative responses address the communication goals indicated by that natural language inputs.
29. The method of claim 28 wherein the data set comprises visualization parameter data.
30. The method of claim 28 wherein the operating parameters comprise a plurality of commands derived from the natural language inputs.
31. The method of claim 30 wherein at least one of the commands comprises a focus task or a drill down task.
32. The method of claim 30 wherein at least one of the commands includes a specification of an entity.
33. The method of claim 32 wherein the entity comprises a specific entity identified in at least one of the natural language inputs.
34. The method of claim 32 wherein the entity comprises a notional entity that is notionally identified in at least one of the natural language inputs.
35. The method of claim 28 further comprising:
the narrative generation artificial intelligence platform (1) deriving information from logic that the narrative generation artificial intelligence platform uses to generate natural language and (2) providing the derived information to the natural language conversational interface; and
the natural language conversational interface defining a language pattern for identifying a pattern in natural language input to the natural language conversational interface based on the provided information.
36. The method of claim 31 wherein the operating parameters include a specification of a notional entity;
wherein the generating step includes the narrative generation artificial intelligence platform determining a natural language expression for referring to an element in at least one of the natural language narrative responses so that the natural language expression serves as the derived information; and
wherein the method further comprises (1) the narrative generation artificial intelligence platform providing the determined natural language expression to the natural language conversational interface and (2) the natural language conversational interface using the provided natural language expression as a language pattern for identifying a reference in at least one of the natural language inputs to the element.
37. The method of claim 36 wherein the element comprises an entity.
38. The method of claim 37 wherein the entity comprises a notional entity.
39. The method of claim 38 wherein the generating step further comprises the narrative generation artificial intelligence platform (1) determining a specific entity to which the notional entity refers based on an analysis of the data set and (2) determining a natural language expression for referring to the specific entity corresponding to the notional entity, and
wherein the method further comprises (1) the narrative generation artificial intelligence platform providing the determined natural language expression for referring to the specific entity corresponding to the notional entity to the natural language conversational interface and (2) the natural language conversational interface using the provided natural language expression for referring to the specific entity corresponding to the notional entity as a language pattern for identifying a reference in at least one of the natural language inputs to the specific entity.
40. The method of claim 31 wherein the derived information comprises information about a command corresponding to at least one of the natural language inputs so that the identified pattern comprises the corresponding command.
41. A system comprising:
a processor configured to interactively generate a narrative story about a data set based on a plurality of processing operations performed with respect to a series of sequentially received natural language inputs from a user, wherein the narrative story comprises a series of unfolding natural language narrative responses to the natural language inputs that form an interactive dialog relating to the data set, wherein each natural language input has a corresponding natural language narrative response and is indicative of a communication goal to be addressed by its corresponding natural language narrative response;
wherein the processing operations include operations where the processor is further configured to, in response to each of the natural language inputs, (1) receive data derived from that natural language input, (2) based on the received data, parameterize a narrative analytics model relevant to the communication goal indicated by that natural language input, (3) execute the parameterized narrative analytics model with respect to the data set to determine content about the data set for an outline of the corresponding natural language narrative response, and (4) generate the corresponding natural language narrative response based on the outline so that the corresponding natural language narrative response addresses the communication goal indicated by that natural language input and expresses the determined content in natural language.
42. The system of claim 41 wherein the processor is part of a narrative generation artificial intelligence platform, the system further comprising:
a natural language conversational interface for cooperation with the narrative generation artificial intelligence platform, wherein the natural language conversational interface is configured to (1) receive the natural language inputs from the user, (2) determine commands for the narrative generation artificial intelligence platform based on the received natural language inputs, and (3) interface with the narrative generation artificial intelligence platform to (i) provide the determined commands to the narrative generation artificial intelligence platform, wherein the provided commands serve as the received data, and (ii) receive the natural language narrative responses from the narrative generation artificial intelligence platform for sequential presentation to the user as part of the interactive dialog.