Matching bias and relevancy in reviews with artificial intelligence

1. A computer-implemented method for bias matching, the method comprising:

gathering, by a number of processors, a review from an internet media source;

parsing, by a number of processors, the review to identify a number of entities within the review;

parsing, by a number of processors, a number of internet media posts to identify entities within the posts;

mapping, by a number of processors, entities in the review to entities in the internet media posts;

determining, by a number of processors, a bias and context of the review, wherein:

if no internet media posts within a specified time frame are mapped to entities in the review, there is no bias in the review;

if a media post within a specified time frame has a bias that does not match a bias in the review, the review is ignored; and

if a media post within a specified time frame has a bias that matches a bias in the review, the review is applicable;

determining, by a number of processors, a bias and context for a user reading the review;

determining, by a number of processors, a relevancy score of the review for the user by comparing the bias and context of the review to the bias and context of the user; and

displaying, by a number of processors, the review among a number of reviews according to its relevancy score for the user.

2. The method of claim 1 , further comprising adjusting weights on bias and context for users and clusters of users according to usage of reviews and feedback from users.

3. The method of claim 1 , wherein the bias and context of the user is derived according to k-mean clustering.

4. The method of claim 1 , wherein the user can manually adjust the relevancy score of the review according to user preference.

5. The method of claim 1 , further comprising storing and analyzing metadata for a number of internet media posts of an author to determine historical trends of bias of the author.

6. The method of claim 1 , wherein matching the bias and context of the review to the bias and context of the user comprises vector matching.

7. The method of claim 1 , wherein the method is implemented in a neural network.

8. A computer program product for bias matching, the computer program product comprising:

a non-volatile computer readable storage medium having program instructions embodied therewith, the program instructions executable by a number of processors to cause the computer to perform the steps of:

gathering a review from an internet media source;

parsing the review to identify a number of entities within the review;

parsing a number of internet media posts to identify entities within the posts;

mapping entities in the review to entities in the internet media posts;

determining a bias and context of the review, wherein:

if no internet media posts within a specified time frame are mapped to entities in the review, there is no bias in the review;

if a media post within a specified time frame has a bias that does not match a bias in the review, the review is ignored; and

if a media post within a specified time frame has a bias that matches a bias in the review, the review is applicable;

determining a bias and context for a user reading the review;

determining a relevancy score of the review for the user by comparing the bias and context of the review to the bias and context of the user; and

displaying the review among a number of reviews according to its relevancy score for the user.

9. The computer program product of claim 8 , further comprising instructions to adjust weights on bias and context for users and clusters of users according to usage of reviews and feedback from users.

10. The computer program product of claim 8 , wherein the bias and context of the user is derived according to k-mean clustering.

11. The computer program product of claim 8 , wherein the user can manually adjust the relevancy score of the review according to user preference.

12. The computer program product of claim 8 , further comprising instructions for storing and analyzing metadata for a number of internet media posts of an author to determine historical trends of bias of the author.

13. The computer program product of claim 8 , wherein matching the bias and context of the review to the bias and context of the user comprises vector matching.

14. The computer program product of claim 8 , wherein the number of processors execute the program instructions within a neural network.

15. A system for bias matching, the system comprising:

a bus system;

a storage device connected to the bus system, wherein the storage device stores program instructions; and

a number of processors connected to the bus system, wherein the number of processors execute the program instructions to:

gather a review from an internet media source;

parse the review to identify a number of entities within the review;

parse a number of internet media posts to identify entities within the posts;

map entities in the review to entities in the internet media posts;

determine a bias and context of the review, wherein:

if no internet media posts within a specified time frame are mapped to entities in the review, there is no bias in the review;

if a media post within a specified time frame has a bias that does not match a bias in the review, the review is ignored; and

if a media post within a specified time frame has a bias that matches a bias in the review, the review is applicable;

determine a bias and context for a user reading the review;

determining a relevancy score of the review for the user by comparing the bias and context of the review to the bias and context of the user; and

display the review among a number of reviews according to its relevancy score for the user.

16. The system according to claim 15 , wherein the number of processors further execute instructions to adjust weights on bias and context for users and clusters of users according to usage of reviews and feedback from users.

17. The system according to claim 15 , wherein the system further comprises a neural network.

18. The method of claim 1 , wherein parsing the internet media posts to identify entities within the posts is performed with natural language processing.

19. The computer program product of claim 8 , wherein parsing the internet media posts to identify entities within the posts is performed with natural language processing.

20. The system of claim 15 , wherein parsing the internet media posts to identify entities within the posts is performed with natural language processing.