Systems and methods for measuring automatability of report generation using a natural language generation system

1. A method for determining a degree to which a document can be generated using a natural language generation (NLG) system, the NLG system being configured to generate natural language text using semantic objects, the method comprising:

using at least one computer hardware processor to perform:

(A) obtaining a document comprising text segments;

(B) determining a degree to which at least some of the text segments can be generated using the NLG system and at least some of the semantic objects;

(C) generating a report indicating the degree to which the at least some of the text segments can be generated using the NLG system and the at least some of the semantic objects; and

(D) outputting the report,

wherein the at least some of the text segments include a first text segment, and wherein (B) comprises:

generating a first annotated representation for the first text segment; and

determining, for each semantic object of one or more of the at least some of the semantic objects, a degree to which the first text segment can be generated by the NLG system using the semantic object, the determining comprising:

accessing a plurality of annotated representations for text segments associated with the semantic object;

determining measures of similarity between the first annotated representation and the plurality of annotated representations; and

determining, using the measures of similarity, the degree to which the first text segment can be generated by the NLG system using the semantic object.

2. The method of claim 1 , wherein the NLG system is configured to generate natural language text for a plurality of topics using a respective plurality of subsets of the semantic objects, wherein each subset of the semantic objects is associated with a respective topic in the plurality of topics, the method further comprising:

processing the document to identify a topic for the document from among the plurality of topics,

wherein (C) comprises:

determining the degree to which the at least some of the text segments in the document can be generated using the NLG system and at least some of the subset of semantic objects associated with the topic for the document.

3. The method of claim 1 , wherein generating the first annotated representation for the first text segment comprises:

tagging words in the first text segment using tags in a plurality of tags.

4. The method of claim 3 , wherein the plurality of tags includes a first tag associated with a first vocabulary in a plurality of vocabularies.

5. The method of claim 4 , wherein tagging words in the first text segment using the tags in the plurality of tags comprises:

determining whether a word in the first text segment is included in a first vocabulary of the plurality of vocabularies; and

when it is determined that the word in the first text segment is in the first vocabulary of the plurality of vocabularies, tagging the word with the first tag.

6. The method of claim 5 , wherein tagging words in the first text segment using the tags in the plurality of tags further comprises:

when it is determined that the word in the first text segment is not included in any of the plurality of vocabularies, determining a part-of-speech for the word; and

tagging the word using a tag associated with the part-of-speech determined for the word.

7. The method of claim 4 , wherein the plurality of vocabularies includes: a vocabulary having words and/or symbols related to currency, a vocabulary having words and/or symbols related to geography, a vocabulary having words and/or symbols related to diseases, a vocabulary having words and/or symbols related to drugs, a vocabulary having words and/or symbols related to financial indicators, a vocabulary having words and/or symbols related to dates, a vocabulary having words and/or symbols related to percentages, and/or a vocabulary having words and/or symbols related to variation.

8. The method of claim 3 , wherein the plurality of annotated representations comprises a second annotated representation, and

wherein determining measures of similarities between the first annotated representation and the plurality of annotated representations comprises:

determining a measure of similarity between the first annotated representation and the second annotated representation, the determining being performed using tags in the first and second annotated representations.

9. The method of claim 8 , wherein determining the measure of similarity between the first annotated representation and the second annotated representation comprises determining a Levenshtein distance among the tags in the first and second annotated representations.

10. The method of claim 1 , wherein determining the degree to which the first text segment can be generated by the NLG system using the semantic object comprises identifying a first measure of similarity from among the measures of similarities, the first measure of similarity being associated with a largest measure of similarity from among the measures of similarities.

11. The method of claim 1 , wherein generating the report comprises generating the report to include information indicative of:

a first semantic object of the one or more of the at least some semantic objects that can be used to generate the first text segment; and

a degree to which the first text segment can be generated using the first semantic object and the NLG system.

12. The method of claim 1 , wherein generating the report comprises generating the report to indicate information indicative of a degree to which one or more regions of the document can be generated using the NLG system and the at least some of the semantic objects.

13. The method of claim 1 , wherein at least some of the text segments includes second and third text segments; and

wherein generating the report comprises generating the report to include information indicative of:

the second text segment when a degree to which the second text segment can be generated using the NLG system and the at least some of the semantic objects exceeds a specified threshold; and

the third text segment when a degree to which the third text segment can be generated using the NLG system and the at least some of the semantic objects does not exceed the specified threshold.

14. The method of claim 1 , wherein outputting the report comprises outputting the report through a graphical user interface (GUI).

15. The method of claim 14 , wherein the GUI is interactive; and

wherein the GUI includes selectable elements through which a user can provide input to associate a particular semantic object to a particular text segment.

16. The method of claim 1 , wherein the semantic objects include a first semantic object associated with one or more data variables, one or more attributes, a vocabulary, and a document structure.

17. The method of claim 1 , wherein the first text segment includes one or more words, a part of a sentence, a sentence, and/or multiple sentences.

18. A system comprising:

at least one computer hardware processor; and

at least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform:

(A) obtaining a document comprising text segments;

(B) determining a degree to which at least some of the text segments can be generated using an NLG system and at least some semantic objects;

(C) generating a report indicating the degree to which the at least some of the text segments can be generated using the NLG system and the at least some of the semantic objects; and

(D) outputting the report,

wherein the at least some of the text segments include a first text segment, wherein (B) comprises:

generating a first annotated representation for the first text segment; and

determining, for each semantic object of one or more of the at least some of the semantic objects, a degree to which the first text segment can be generated by the NLG system using the semantic object, the determining comprising:

accessing a plurality of annotated representations for text segments associated with the semantic object;

determining measures of similarity between the first annotated representation and the plurality of annotated representations; and

determining, using the measures of similarity, the degree to which the first text segment can be generated by the NLG system using the semantic object.

19. At least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform:

(A) obtaining a document comprising text segments;

(B) determining a degree to which at least some of the text segments can be generated using an NLG system and at least some semantic objects;

(C) generating a report indicating the degree to which the at least some of the text segments can be generated using the NLG system and the at least some of the semantic objects; and

(D) outputting the report,

wherein the at least some of the text segments include a first text segment, wherein (B) comprises:

generating a first annotated representation for the first text segment; and

determining, for each semantic object of one or more of the at least some of the semantic objects, a degree to which the first text segment can be generated by the NLG system using the semantic object, the determining comprising:

accessing a plurality of annotated representations for text segments associated with the semantic object;

determining measures of similarity between the first annotated representation and the plurality of annotated representations; and

determining, using the measures of similarity, the degree to which the first text segment can be generated by the NLG system using the semantic object.