Initializing a workspace for building a natural language understanding system

1. A method for initializing a workspace for building a Natural Language Understanding (NLU) system, the method comprising:

generating a plurality of prompt match scores of at least one candidate stub grammar associated with a sentence of a plurality of expected sentences by comparing a sample sentence to the plurality of expected sentences associated with the at least one candidate stub grammar, each prompt match score based on a semantic similarity of the sample sentence to each one of the plurality of expected sentences;

automatically selecting sets of at least one candidate stub grammar model from among a library of multiple candidate stub grammar models stored in a memory, the automatically selected sets of at least one candidate stub grammar model being associated with expected sentences having prompt match scores above a particular threshold, each candidate stub grammar model being a model correlating an input to a response, each candidate stub grammar further being created prior to the comparing of the sample sentence to the plurality of expected sentences, the model being associated with at least one concept associated with the sample sentence;

presenting, to a user, respective representations of the automatically selected sets of at least one candidate stub grammar model at a display;

enabling the user to choose one of the respective representations of the candidate stub grammar models presented on the display, the chosen one of the respective representations being associated with a chosen candidate stub grammar; and

adding, using the processor, to the workspace the chosen stub grammar model corresponding to the representation of the candidate stub grammar model chosen by the user, the workspace being an ontology configured to store a plurality of stub grammar models, the workspace configured to build the NLU system.

2. The method of claim 1 , wherein the respective representations are names of the candidate stub grammar models.

3. The method of claim 1 , further comprising presenting each respective representation of the candidate stub grammar model to the user by visually presenting each of the respective representations of the candidate stub grammar model relative to a corresponding part of the sample sentence.

4. The method of claim 1 , wherein the presented respective representations have a semantic match to the parsed sample sentence above a threshold.

5. The method of claim 1 , wherein parsing the sample sentence further includes comparing the sample sentence to a stored sentence that is associated with at least one candidate stub grammar model, and if the sample sentence is semantically related to the stored sentence, selecting the at least one candidate stub grammar model associated with the stored sentence.

6. The method of claim 1 , wherein parsing the sample sentence includes parsing the sample sentence to generate a set of annotations including mentions.

7. The method of claim 6 , wherein the annotations further include mentions and intents.

8. A system for initializing a workspace for building a Natural Language Understanding (NLU) system, the system comprising:

a parsing module configured to:

generate a plurality of prompt match scores of at least one candidate stub grammar associated with a sentence of the plurality of expected sentences by comparing a sample sentence to the plurality of expected sentences associated with the at least one candidate stub grammar, each prompt match score based on a semantic similarity of the sample sentence to each one of the plurality of expected sentences, and

automatically select sets of at least one candidate stub grammar model from among a library of multiple candidate stub grammar models stored in a memory, the automatically selected sets of at least one candidate stub grammar model being associated with expected sentences having prompt match scores above a particular threshold, each candidate stub grammar model being a model correlating an input to a response, each candidate stub grammar further being created prior to the comparing of the sample sentence to the plurality of expected sentences, the model being associated with at least one concept associated with the sample sentence;

a presentation module configured to present, at a display to a user, respective representations of the at least one automatically selected sets of candidate stub grammar model;

a user interface module configured to enable the user to choose one of the respective representations of the candidate stub grammar models presented on the display, the chosen one of the respective representations being associated with a chosen candidate stub grammar; and

a workspace building module configured to add, using the processor, the chosen grammar model, chosen by the user from among the respective presented representations, to the workspace, the workspace being an ontology configured to store a plurality of stub grammar models, the workspace configured to build the NLU system.

9. The system of claim 8 , wherein the respective representations are names of the candidate stub grammar models.

10. The system of claim 8 , wherein the presentation module is further configured to visually present each of the respective representations of the candidate stub grammar models relative to a corresponding part of the sample sentence.

11. The system of claim 8 , wherein the presented respective representations have a semantic match to the parsed sample sentence above a threshold.

12. The system of claim 8 , wherein the parsing module is further configured to parse the sample sentence to generate a set of annotations including mentions.

13. The system of claim 12 , wherein the annotations further include mentions and intents.

14. A non-transitory computer-readable medium configured to store instructions for initializing a workspace for building a Natural Language Understanding (NLU) system, the instructions, when loaded and executed by a processor, cause the processor to:

generate a plurality of prompt match scores of at least one candidate stub grammar associated with a sentence of the plurality of expected sentences by comparing a sample sentence to the plurality of expected sentences associated with the at least one candidate stub grammar, each prompt match score based on a semantic similarity of the sample sentence to each one of the plurality of expected sentences, and

automatically select sets of at least one candidate stub grammar model from among a library of multiple candidate stub grammar models stored in a memory, the automatically selected sets of at least one candidate stub grammar model being associated with expected sentences having prompt match scores above a particular threshold, each candidate stub grammar model being a model correlating an input to a response, each candidate stub grammar further being created prior to the comparing of the sample sentence to the plurality of expected sentences, the model being associated with at least one concept associated with the sample sentence;

present, on a display to a user, respective representations of the automatically selected sets of at least one candidate stub grammar model;

enable the user to choose one of the respective representations of the candidate stub grammar models presented on the display, the chosen one of the respective representations being associated with a chosen candidate stub grammar; and

add, using the processor, to the workspace the chosen stub grammar model corresponding to the representation of the candidate stub grammar model chosen by the user, the workspace being an ontology configured to store a plurality of stub grammar models, the workspace configured to build the NLU system.

15. The non-transitory computer-readable medium of claim 14 , wherein the respective representations are names of the candidate stub grammar models.

16. The non-transitory computer-readable medium of claim 14 , wherein the instructions further cause the processor to present each respective representation of the candidate stub grammar models to the user by visually presenting each of the respective representations of the candidate stub grammar models relative to a corresponding part of the sample sentence.

17. The non-transitory computer-readable medium of claim 14 , wherein the presented respective representations have a semantic match to the parsed sample sentence above a threshold.

18. The non-transitory computer-readable medium of claim 14 , wherein instructions further cause the processor to parse the sample sentence to generate a set of annotations including mentions.

19. The non-transitory computer-readable medium of claim 18 , wherein the annotations further include mentions and intents.