Method and system for suggesting revisions to an electronic document

1. A method for suggesting revisions to text data, the method comprising:

obtaining a text-under-analysis (“TUA”);

obtaining a candidate original text from a plurality of original texts;

identifying a first edit operation of the candidate original text with respect to a candidate final text associated with the candidate original text, the first edit operation having an edit-type classification;

selecting an alignment method from a plurality of alignment methods based on the edit-type classification of the first edit operation;

identifying a second edit operation based on the selected alignment method; and

creating an edited TUA (“ETUA”) by applying to the TUA the second edit operation.

2. The method of claim 1 , wherein obtaining the candidate original text from the plurality of original text further comprises:

obtaining a first original text from the plurality of original texts;

identifying a first edit operation of the first original text with respect to a first final text associated with the first original text, the first edit operation having a first edit-type classification;

generating a first similarity score for the first original text based on the first edit-type classification, the first similarity score representing a degree of similarity between the TUA and the first original text;

obtaining a second original text from the plurality of original texts;

identifying a second edit operation of the second original text with respect to a second final text associated with the second original text, the second edit operation having a second edit-type classification;

generating a second similarity score for the second original text based on the second edit-type classification, the second similarity score representing a degree of similarity between the TUA and the second original text; and,

selecting the candidate original text from one of the first original text and the second original text.

3. The method of claim 1 , wherein the edit-type classification comprises a point edit, and the selected alignment method comprises:

aligning the TUA with the candidate original text; and

determining a set of one or more edit operations to modify the TUA consistent with the first candidate final text associated with the first candidate original text, wherein the set of one or more edit operations comprises the second edit operation.

4. The method of claim 1 , wherein the selected alignment method comprises:

aligning a portion of the TUA with a portion of the candidate original text that is lexically different than the portion of the TUA based on a semantic similarity between the portion of the TUA and the portion of the candidate original text.

5. The method of claim 1 , wherein the first edit operation is a deletion edit operation having a full edit type, and wherein the selected alignment method comprises:

selecting a set of original texts from the plurality of original texts, wherein each original text in the set of original texts has a similarity score representing a degree of similarity between the TUA and the original text;

creating a subset of original texts from the selected set of original texts, wherein each original text in the subset has a deletion edit operation having a full edit type classification with respect to each final text associated with each original text in the subset; and

determining that the second edit operation is a deletion of the TUA based on the subset of original texts.

6. The method of claim 1 , wherein the edit-type classification is a full edit type, and the alignment method comprises:

obtaining a set of final texts, wherein each final text in the set is associated with an original text having a full edit type classification;

obtaining a set of candidate edit operations of the full edit type classification from the set of final texts;

selecting the second edit operation from the set of candidate edit operations, wherein the second edit operation is associated with a candidate final text; and

aligning one or more features of the TUA with one or more features of the candidate final text.

7. The method of claim 6 , wherein the second edit operation is an insertion, the method further comprising:

determining that a text corresponding to the insertion edit operation is not located in the TUA or in a document under analysis comprising the TUA.

8. The method of claim 6 , further comprising:

determining that a first portion of the text corresponding to the insertion edit operation is not located in the TUA or the document under analysis comprising the TUA; and,

determining that a second portion of the text corresponding to the insertion edit operation is located in the TUA or the document under analysis comprising the TUA, wherein

the second edit operation comprises inserting the first portion of text and not inserting the second portion of text.

9. The method of claim 1 , wherein the edit-type classification is a span edit, and the alignment method comprises:

aligning a portion of the TUA with a portion of the candidate original text.

10. The method of claim 1 , wherein the edit-type classification is a list edit, and the alignment method comprises:

tokenizing the TUA into a plurality of list items under analysis;

tokenizing the candidate final text into a plurality of candidate list items;

aligning each list item under analysis with each candidate list item; and

determining, based on the aligning, that the second edit operation comprises an insertion or deletion of a list item to the TUA.

11. A non-transitory computer readable medium storing instructions configured to cause a computer to perform the method of claim 1 .

12. A system for suggesting revisions to text data, the system comprising:

a processor;

a non-transitory computer readable memory coupled to the processor, wherein the processor is configured to:

obtain a text-under-analysis (“TUA”);

obtain a candidate original text from a plurality of original texts;

identify a first edit operation of the candidate original text with respect to a candidate final text associated with the candidate original text, the first edit operation having an edit-type classification;

select an alignment method from a plurality of alignment methods based on the edit-type classification of the first edit operation;

identify a second edit operation based on the selected alignment method; and

create an edited TUA (“TSUA”) by applying to the TUA the second edit operation.

13. The system of claim 12 , wherein the processor is further configured to:

obtain a first original text from the plurality of original texts;

identify a first edit operation of the first original text with respect to a first final text associated with the first original text, the first edit operation having a first edit-type classification;

generate a first similarity score for the first original text based on the first edit-type classification, the first similarity score representing a degree of similarity between the TUA and the first original text;

obtain a second original text from the plurality of original texts;

identify a second edit operation of the second original text with respect to a second final text associated with the second original text, the second edit operation having a second edit-type classification;

generate a second similarity score for the second original text based on the second edit-type classification, the second similarity score representing a degree of similarity between the TUA and the second original text; and,

select the candidate original text from one of the first original text and the second original text.

14. The system of claim 12 , wherein the edit-type classification is a point edit, and the processor is further configured to:

align the TUA with the candidate original text; and

determine a set of one or more edit operations to modify the TUA consistent with the first candidate final text associated with the first candidate original text, wherein the set of one or more edit operations comprises the second edit operation.

15. The system of claim 12 , wherein the processor is further configured to:

align a portion of the SUA with a portion of the candidate original text that is lexically different than the portion of the SUA based on a semantic similarity between the portion of the SUA and the portion of the candidate original text.

16. The system of claim 12 , wherein the first edit operation is a deletion edit operation having a full edit type, and wherein the processor is further configured to:

select a set of original texts from the plurality of original texts, wherein each original text in the set of original texts has a similarity score representing a degree of similarity between the TUA and the original text;

create a subset of original texts from the set of original texts, wherein each original text in the subset has a deletion edit operation having a full edit type with respect to each final text associated with each original text in the subset; and

determine that the second edit operation is a deletion of the TUA based on the on the subset of original texts.

17. The system of claim 12 , wherein the edit-type classification is a full edit type, and wherein the processor is further configured to:

obtain a set of final texts, wherein each final text in the set is associated with an original text having a full edit type classification;

obtain a set of candidate edit operations of the full edit type classification from the set of final texts;

select the second edit operation from the set of candidate edit operations, wherein the second edit operation is associated with a candidate final text; and

align one or more features of the TUA with one or more features of the candidate final text.

18. The system of claim 17 , wherein the second edit operation is an insertion and wherein processor is further configured to:

determine that a text corresponding to the insertion is not located in the TUA or in a document under analysis comprising the TUA.

19. The system of claim 12 , wherein the edit-type classification is a span edit and wherein the processor is further configured to:

align a portion of the TUA with a portion of the candidate original text.

20. The system of claim 12 , wherein the edit-type classification is a list edit and wherein the processor is further configured to:

tokenize the TUA into a plurality of list items under analysis;

tokenize the candidate final text into a plurality of candidate list items;

align each list item under analysis with each candidate list item; and

determine, based on the alignment, that the second edit operation comprises an insertion or deletion of a list item to the TUA.