Suggestions on removing cognitive terminology in news articles

Patent metadata

Field	Value
Publication number	US-11074417-B2
Application number	US-201916264015-A
Country	US
Kind code	B2
Filing date	Jan 31, 2019
Priority date	Jan 31, 2019
Publication date	Jul 27, 2021
Grant date	Jul 27, 2021

Abstract

Embodiments provide a computer implemented method in a data processing system comprising a processor and a memory comprising instructions, which are executed by the processor to cause the processor to implement the method of removing a cognitive terminology from a news article at a news portal, the method including: receiving, by the processor, a first news article from a user; configuring, by the processor, a cognitive terminology filter list to add one or more entities and one or more cognitive terminology types associated with each entity in the cognitive terminology filter list; dividing, by the processor, the first news article into a plurality of text segments; identifying, by the processor, one or more key entities and one or more inter-entity relationships of each text segment; detecting, by the processor, one or more cognitive terminologies in the first news article; and providing, by the processor, one or more suggestions to remove the one or more cognitive terminologies.

First claim

What is claimed is:

1. A computer implemented method in a data processing system comprising a processor and a memory comprising instructions, which are executed by the processor to cause the processor to implement the method of removing a bias from a news article at a news portal, the method comprising:
receiving, by the processor, a first news article from a user;
configuring, by the processor, a bias filter list to add one or more entities and a stance associated with each entity in the bias filter list, wherein the stance is “positive” or “negative”;
dividing, by the processor, the first news article into a plurality of text segments;
identifying, by the processor, one or more key entities and one or more inter-entity relationships of each text segment;
detecting, by the processor, one or more biases in the first news article;
ignoring, by the processor, a stance bias of the one or more biases if a particular entity and a stance associated with the particular entity are included in the bias filter list; and
providing, by the processor, one or more suggestions to remove the one or more biases.

2. The method as recited in claim 1 , wherein the step of detecting further comprising:
calculating, by the processor, a stance score of each text segment;
calculating, by the processor, a sentiment score of each text segment;
calculating, by the processor, a tone score of each text segment;
obtaining, by the processor, a bias score of each text segment based on the stance score, the sentiment score, and the tone score; and
identifying, by the processor, the one or more biases if the bias score of a particular text segment is over a predetermined threshold value.

3. The method as recited in claim 2 , wherein the step of obtaining further comprising:
summing, by the processor, the stance score, the sentiment score, and the tone score to obtain a bias score of each text segment, wherein the stance score, the sentiment score, and the tone score are each assigned with a different weight.

4. The method as recited in claim 2 , further comprising:
calculating the stance score, the sentiment score, and the tone score of each text segment based on one or more of the first news article of the user, other news articles created by the user prior to the first news article, other news articles written by other users at the same news portal, and other news articles created by other users at other news portals.

5. The method as recited in claim 4 , further comprising:
retrieving the other news articles created by the user prior to the first news article, the other news articles written by the other users at the same news portal, and the other news articles created by the other users at the other news portals, through a web crawler integrated in a search engine.

6. The method as recited in claim 1 , further comprising:
providing, by the processor, additional suggestions to avoid creating new biases in the first news article.

7. A computer program product for removing a bias from a news article at a news portal, 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 processor to:
receive a first news article from a user;
configure a bias filter list to add one or more entities and a stance associated with each entity in the bias filter list, wherein the stance is “positive” or “negative”;
divide the first news article into a plurality of text segments;
identify one or more key entities and one or more inter-entity relationships of each text segment;
detect one or more biases in the first news article;
ignore a stance bias of the one or more biases if a particular entity and a stance associated with the particular entity are included in the bias filter list; and
provide one or more suggestions to remove the one or more biases.

8. The computer program product as recited in claim 7 , wherein the processor is further caused to:
calculate a stance score of each text segment;
calculate a sentiment score of each text segment;
calculate a tone score of each text segment;
obtain a bias score of each text segment based on the stance score, the sentiment score, and the tone score; and
identify the one or more biases if the bias score of a particular text segment is over a predetermined threshold value.

9. The computer program product as recited in claim 8 , wherein the processor is further caused to:
calculate the stance score, the sentiment score, and the tone score of each text segment based on one or more of the first news article of the user, other news articles created by the user prior to the first news article, other news articles written by other users at the same news portal, and other news articles created by other users at other news portals.

10. The computer program product as recited in claim 9 , wherein the processor is further caused to:
retrieve the other news articles created by the user prior to the first news article, the other news articles written by the other users at the same news portal, and the other news articles created by the other users at the other news portals, through a web crawler integrated in a search engine.

11. The computer program product as recited in claim 7 , wherein the processor is further caused to:
provide the one or more suggestions based on a plurality of news articles from an internet through a web crawler integrated in a search engine.

12. The computer program product as recited in claim 7 , wherein the processor is further caused to:
provide the one or more suggestions using a generative network.

13. A system for removing a bias from a news article at a news portal, comprising:
a processor configured to:
receive a first news article from a user;
configure a bias filter list to add one or more entities and a stance associated with each entity in the bias filter list, wherein the stance is “positive” or “negative”;
divide the first news article into a plurality of text segments;
identify one or more key entities and one or more inter-entity relationships of each text segment;
detect one or more biases in the first news article;
ignore a stance bias of the one or more biases if a particular entity and a stance associated with the particular entity are included in the bias filter list; and
provide to the user one or more suggestions to remove the one or more biases.

14. The system as recited in claim 13 , the processor is further configured to:
calculate a stance score of each text segment;
calculate a sentiment score of each text segment;
calculate a tone score of each text segment;
obtain a bias score of each text segment based on the stance score, the sentiment score, and the tone score; and
identify the one or more biases if the bias score of a particular text segment is over a predetermined threshold value.

15. The system as recited in claim 14 , wherein the processor is further configured to:
calculate the stance score, the sentiment score, and the tone score of each text segment based on one or more of the first news article of the user, other news articles created by the user prior to the first news article, other news articles written by other users at the same news portal, and other news articles created by other users at other news portals.

16. The system as recited in claim 15 , wherein the processor is further configured to:
provide the one or more suggestions based on a plurality of news articles from an internet through a web crawler integrated in a search engine.

Assignees

• IBM

Inventors

• Bhide Manish A
• Mehta Sameep
• Madaan Nishtha
• Dey Kuntal

Classifications

• G06F40/289 Primary

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• G06F16/9536

  Search customisation based on social or collaborative filtering · CPC title

• G06F40/35 Primary

  Discourse or dialogue representation · CPC title

• G06F40/169

  Annotation, e.g. comment data or footnotes · CPC title