Systems and methods for automatically creating a patent application based on a claim set such that the patent application follows a document plan inferred from an example document
Systems and methods for automatically creating a patent application based on a claim set such that the patent application follows a document plan inferred from an example document are disclosed. Exemplary implementations may: obtain one or more example documents, a given example document including a patent document; identify common formatting features among the one or more example documents; determine a document plan for a patent application based on information gained from the one or more example documents; receive one or more previously unseen patent claims; and use computerized natural language generation to automatically create a new patent application based on both the document plan and the one or more patent claims such that the new patent application reflects subject matter of the one or more patent claims and is otherwise consistent with the one or more example documents.
1. A system configured for automatically creating a patent application based on a claim set such that the patent application follows a document plan inferred from example documents, the system comprising:
one or more hardware processors configured by machine-readable instructions to:
obtain three or more example documents, a given example document including a patent document, wherein the patent document includes an example patent application or an issued patent;
identify common formatting features among the three or more example documents, wherein the common formatting features include formatting features found in a majority of the three or more example documents;
determine a document plan for a patent application based on information gained from the three or more example documents, the information including the common formatting features, common language, and generalized alignment between claim features and specification disclosure;
receive one or more previously unseen patent claims, the one or more previously unseen patent claims being absent from the three or more example documents; and
use computerized natural language generation to automatically create a new patent application based on both the document plan and the one or more previously unseen patent claims such that the new patent application reflects subject matter of the one or more previously unseen patent claims and is otherwise consistent with the three or more example documents.
2. The system of claim 1 , wherein the common formatting features include one or more of font size, font style, line spacing, paragraph numbering style, line numbering style, page numbering style, margin width, or claim type.
3. The system of claim 1 , wherein the common formatting features are extracted from the three or more example documents.
4. The system of claim 1 , wherein the one or more hardware processors are further configured by machine-readable instructions to identify the common language among the three or more example documents, wherein the common language includes one or more of header text, footer text, boilerplate language, or common disclosure.
5. The system of claim 4 , wherein the common language is indicative of static and variable portions of the three or more example documents.
6. The system of claim 4 , wherein the common language is identified by determining a word-level edit distance between pairs of sentences or paragraphs.
7. The system of claim 1 , wherein the one or more hardware processors are further configured by machine-readable instructions to determine the generalized alignment between claim features and corresponding disclosure in a specification of a same patent document, wherein the generalized alignment conveys how content of a parsed claim are organized in the specification.
8. The system of claim 1 , wherein the document plan conveys one or more of a general document layout, a document format, content of a document, content organization within a document, or surface text of a document.
9. The system of claim 1 , wherein the one or more previously unseen patent claims are obtained via a same email as the three or more example documents.
10. A method to automatically create a patent application based on a claim set such that the patent application follows a document plan inferred from example documents, the method comprising:
obtaining three or more example documents, a given example document including a patent document, wherein the patent document includes an example patent application or an issued patent;
identifying common formatting features among the three or more example documents, wherein the common formatting features include formatting features found in a majority of the three or more example documents;
determining a document plan for a patent application based on information gained from the three or more example documents, the information including the common formatting features, common language, and generalized alignment between claim features and specification disclosure;
receiving one or more previously unseen patent claims, the one or more previously unseen patent claims being absent from the three or more example documents; and
using computerized natural language generation to automatically create a new patent application based on both the document plan and the one or more previously unseen patent claims such that the new patent application reflects subject matter of the one or more previously unseen patent claims and is otherwise consistent with the three or more example documents.
11. The method of claim 10 , wherein the common formatting features include one or more of font size, font style, line spacing, paragraph numbering style, line numbering style, page numbering style, margin width, or claim type.
12. The method of claim 10 , wherein the common formatting features are extracted from the three or more example documents.
13. The method of claim 10 , further comprising identifying the common language among the three or more example documents, wherein the common language includes one or more of header text, footer text, boilerplate language, or common disclosure.
14. The method of claim 13 , wherein the common language is indicative of static and variable portions of the three or more example documents.
15. The method of claim 13 , wherein the common language is identified by determining a word-level edit distance between pairs of sentences or paragraphs.
16. The method of claim 10 , further comprising determining the generalized alignment between claim features and corresponding disclosure in a specification of a same patent document, wherein the generalized alignment conveys how content of a parsed claim are organized in the specification.
17. The method of claim 10 , wherein the document plan conveys one or more of a general document layout, a document format, content of a document, content organization within a document, or surface text of a document.
18. The method of claim 10 , wherein the one or more previously unseen patent claims are obtained via a same email as the three or more example documents.