Real-time assessment of text consistency

Patent metadata

Field	Value
Publication number	US-2020302011-A1
Application number	US-201916361701-A
Country	US
Kind code	A1
Filing date	Mar 22, 2019
Priority date	Mar 22, 2019
Publication date	Sep 24, 2020
Grant date	—

Abstract

Text suggestions are generated. A document is received, and a portion score for at least one portion of the document is generated. A global assessment score for at least two portions of the document is also generated. A variation between the portion score and the global assessment score is calculated. It is determined that the variation is above a threshold variation, and at least one text change suggestion is generated.

First claim

What is claimed is:

1 . A method of generating text change suggestions, comprising:
receiving a document; generating a portion score for at least one portion of the document; generating a global assessment score for at least two portions of the document; calculating a variation between the portion score and the global assessment score; determining that the variation is above a threshold variation; and generating at least one text change suggestion.

2 . The method of claim 1 , wherein the global assessment score and the portion score are based on semantic artifacts.

3 . The method of claim 2 , wherein the semantic artifacts include at least one semantic artifact selected from the group consisting of bias-based semantic artifacts, correctness-based semantic artifacts, and coherence-based semantic artifacts.

4 . The method of claim 1 , wherein the global assessment score and the portion score are based on semantic artifacts and contributor scores.

5 . The method of claim 1 , wherein the at least one text change suggestion is generated using style-based transformers.

6 . The method of claim 1 , further comprising:
identifying at least one leader from at least two contributors to the document; and generating the global assessment score based on scores for portions of the document contributed by the at least one leader.

7 . The method of claim 1 , wherein the at least one text change suggestion indicates that the at least one portion contains information that contradicts information in at least one portion of the at least two portions of the document.

8 . A system, comprising:
at least one processing component; at least one memory component; a display screen configured to display a document; a dictionary; and an editing module, comprising:
an assessment module configured to:
receive a document;
generate a portion score for at least one portion of the document;
generate a global assessment score for at least two portions of the document; and
calculate a variation between the portion score and the global assessment score; and

a suggestion module configured to generate at least one text change suggestion.

9 . The system of claim 8 , wherein the global assessment scores and the portion scores are based on semantic artifacts.

10 . The system of claim 9 , wherein the semantic artifacts include at least one semantic artifact selected from the group consisting of bias-based semantic artifacts, correctness-based semantic artifacts, and coherence-based semantic artifacts.

11 . The system of claim 8 , global assessment score and the portion score are based on semantic artifacts and contributor scores.

12 . The system of claim 8 , wherein the assessment module is further configured to:
identify at least one leader from at least two contributors to the document; and generate the global assessment score based on scores for portions of the document contributed by the at least one leader.

13 . The system of claim 8 , wherein the at least one text change suggestion indicates that the at least one portion contains information that contradicts information in at least one portion of the at least two portions of the document.

14 . The system of claim 8 , further comprising a set of contributor profiles.

15 . A computer program product for generating text change suggestions, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the device to perform a method, the method comprising:
receiving a document; generating a portion score for at least one portion of the document; generating a global assessment score for at least two portions of the document; calculating the variation between the portion score and the global assessment score; determining that the variation is above a threshold variation; and generating at least one text change suggestion.

16 . The computer program product of claim 15 , wherein the global assessment score and the portion score are based on semantic artifacts.

17 . The computer program product of claim 16 , wherein the semantic artifacts include at least one semantic artifact selected from the group consisting of bias-based semantic artifacts, correctness-based semantic artifacts, and coherence-based semantic artifacts.

18 . The computer program product of claim 15 , global assessment score and the portion score can also be based on semantic artifacts and contributor scores.

19 . The computer program product of claim 15 , wherein the at least one text change suggestion is generated using style-based transformers.

20 . The computer program product of claim 15 , further comprising:
identifying at least one leader from at least two contributors to the document; and generating the global assessment score based on scores for portions of the document contributed by the at least one leader.

Assignees

• IBM

Inventors

• Mishra Abhijit
• Laha Anirban
• Jain Parag
• Sankaranarayanan Karthik

Classifications

• G06F40/253 Primary

  Grammatical analysis; Style critique · CPC title

• G06F40/242

  Dictionaries · CPC title

• G06F40/166

  Editing, e.g. inserting or deleting · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F16/3322

  using system suggestions (G06F16/3325 takes precedence) · CPC title