Method and apparatus for prediction of community reaction to a post

• US 9,852,239 B2
• Filed: 09/24/2012
• Issued: 12/26/2017
• Est. Priority Date: 09/24/2012
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Natarajan, Anandhavelu, Srinivasan, Balaji Vasan, Gupta, Vineet, Ganesan, Anand, Jain, Anuj, Revankar, Shriram, Singh, Japnik, Polineni, Bharat
• Primary Examiner(s): Rivas, Omar Fernandez
• Assistant Examiner(s): Luu, Cuong
• Application Number:
  US13/625,193
• Publication Number:
  US 20140088944A1
• Time in Patent Office: 1,919 Days
• Field of Search: None
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A method and apparatus for prediction of community reaction to a post for an online social community is disclosed. The method comprises receiving a proposed post as input to a generated prediction model prior to the proposed post being posted to an online social community; predicting a community reaction to the proposed post using the prediction model; and displaying the predication, wherein the prediction comprises a sentiment score and at least one of a number of responses, a number of responders to the post, a longevity of the post, or a half-life of the post.

Claims

1. A computer-implemented method for predicting online social community reaction comprising:

receiving a proposed post prior to the proposed post being posted to an online social community;

generating, by at least one processor, a predicted community reaction to the proposed post;

providing the predicted community reaction prior to the proposed post being posted to the online social community by displaying a plurality of words of the proposed post that contribute to the predicted community reaction, wherein a color of one or more words from the plurality of words reflects a performance indicator and wherein a font size of the one or more words indicates extremity of a predicted online social community reaction corresponding to the one or more words, and wherein the predicted community reaction comprises a predicted sentiment score and at least one of a predicted number of responses, a predicted number of responders to the proposed post, a predicted longevity of the proposed post, or a predicted half-life of the proposed post;

suggesting, by the at least one processor, one or more revisions to the proposed post; and

posting the proposed post, with the suggested one or more revisions, to the online social community.

2. The method of claim 1, wherein generating the predicted community reaction comprises:

gathering, by the at least one processor, a plurality of posts from one or more online social communities and at least one response to each post of the plurality of posts;

calculating, by the at least one processor, a plurality of performance indicators for each post in the plurality of posts, wherein the plurality of performance indicators comprise a sentiment score for the at least one response and at least one of a number of responses to the post, a number of responders to the post, a longevity of the post, or a half-life of the post;

de-biasing the calculated plurality of performance indicators; and

building, by the at least one processor, a prediction model using the de-biased plurality of performance indicators.

3. The method of claim 1, further comprising:

identifying new terms based on the proposed post and the predicted community reaction;

predicting a new sentiment score for each of the new terms; and

suggesting the one or more revisions to the proposed post based on the predicted sentiment score and the new sentiment score for each of the new terms, wherein the suggested one or more revisions comprises at least one new term from the new terms.

4. The method of claim 2, wherein de-biasing comprises calculating a mean (m) and a standard deviation (s) of the plurality of performance indicators for a given post from the plurality of posts and calculating a normalized score (r) to determine a de-biased score r′

, where r′

=(r−

m)/s.

5. The method of claim 2, wherein building the prediction model comprises:

accessing the plurality of posts and a score for each post in the plurality of posts;

retrieving a plurality of terms from each post in the plurality of posts;

calculating a conditional probability for each term in the plurality of terms;

determining one or more terms with a conditional probability that maximize entropy; and

adding the determined one or more terms to the prediction model.

6. The method of claim 1, wherein providing the predicted community reaction further comprises:

displaying a plurality of bins corresponding to at least a portion of the predicted community reaction, wherein a color of one or more bins indicates confidence in the at least a portion of the predicted community reaction corresponding to the one or more bins.

7. The method of claim 1, wherein generating the predicted community reaction comprises:

identifying a plurality of terms and bigrams from the proposed post;

determining a class for which a prediction is to be made for one or more terms and bigrams in the plurality of identified terms and bigrams; and

determining a probability that the one or more terms and bigrams in the plurality of identified terms and bigrams will be in the class.

8. The method of claim 4, wherein predicting the new sentiment score for each of the new terms further comprises:

identifying at least one term from the proposed post with a negative predicted sentiment score; and

predicting sentiment scores for synonyms of the at least one term with the negative predicted sentiment score.

9. A system for predicting an online social community reaction prior to sharing a post comprising:

at least one processor; and

at least one non-transitory computer readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the system to;

receive a proposed post as input to a prediction model prior to the proposed post being posted to an online social community;

generate, by the at least one processor, a predicted community reaction to the proposed post using the prediction model;

provide the predicted community reaction prior to the proposed post being posted to the online social community by displaying a plurality of words of the proposed post that contribute to the predicted community reaction, wherein a color of one or more words from the plurality of words reflects a performance indicator and wherein a font size of the one or more words indicates extremity of a predicted online social community reaction corresponding to the one or more words, and wherein the predicted community reaction comprises a predicted sentiment score and at least one of a predicted number of responses, a predicted number of responders to the proposed post, a predicted longevity of the proposed post, or a predicted half-life of the proposed post;

post the proposed post, with the suggested one or more revisions, to the online social community.

10. The system of claim 9, further comprising instructions thereon, that when executed by the at least one processor, cause the system to:

gather, by the at least one processor, a plurality of posts from one or more online social communities and at least one response to each post of the plurality of posts;

calculate, by the at least one processor, a plurality of performance indicator for the at least one response, wherein the plurality of performance indicators comprise a sentiment score for the at least one response and at least one of a number of responses to the post, a number of responders to the post, a longevity of the post, a sentiment score for the at least one response or a half-life of the post;

de-bias the calculated plurality of performance indicators; and

build the prediction model using the de-biased plurality of performance indicators.

11. The system of claim 9, wherein providing the predicted community reaction comprises:

displaying a plurality of bins corresponding to at least a portion of the predicted community reaction, wherein a color of one or more bins indicates confidence in the at least a portion of the predicted community reaction corresponding to the one or more bins.

12. The system of claim 10, wherein de-biasing comprises calculating a mean (m) and a standard deviation (s) of the plurality of performance indicators for responses to a given post and calculating a normalized score (r) to determine a de-biased score r′

, where r′

=(r−

m)/s.

13. The system of claim 10, wherein building the prediction model comprises:

accessing the plurality of posts and a score for each post in the plurality of posts;

retrieving a plurality of terms from each post in the plurality of posts;

calculating a conditional probability for each term in the plurality of terms;

determining one or more terms with a conditional probability that maximize entropy; and

adding the determined one or more terms to the prediction model.

14. The system of claim 9, wherein generating a predicted community reaction comprises:

identifying a plurality of terms and bigrams from the proposed post;

determining a class for which a prediction is to be made for one or more terms and bigrams in the plurality of identified terms and bigrams; and

determining a probability that the one or more terms and bigrams in the plurality of identified terms and bigrams will be in the class.

15. The system of claim 9, further comprising instructions, that when executed by the at least one processor, cause the system to:

identify new terms based on the proposed post and the predicted community reaction;

predict a new sentiment score for each of the new terms; and

suggest one or more revisions to the proposed post based on the predicted sentiment score and the new sentiment score for each of the new terms, wherein the suggested one or more revisions comprises at least one new term from the new terms.

16. A method for predicting community reaction to a post comprising:

receiving a proposed post as input to a prediction model prior to the proposed post being posted to an online social community;

generating, by at least one processor, a predicted community reaction to the proposed post using the prediction model;

providing the predicted community reaction prior to the proposed post being posted to the online social community, wherein the predicted community reaction comprises a predicted sentiment score and at least one of a predicted number of responses, a predicted number of responders to the proposed post, a predicted longevity of the proposed post, or a predicted half-life of the proposed post by;

displaying a plurality of bins corresponding to at least a portion of the predicted community reaction, wherein each of the plurality of bins reflects an interval of at least one of sentiment scores, number of responses, number of responders, longevity, or half-life, and wherein a color of one or more bins indicates confidence in the at least a portion of the predicted community reaction corresponding to the one or more bins; and

displaying a plurality of words of the proposed post that contribute to the predicted community reaction, wherein a color of one or more words from the plurality of words reflects a performance indicator and wherein a font size of the one or more words indicates extremity of a predicted online social community reaction corresponding to the one or more words;

suggesting, by the at least one processor, one or more revisions to the proposed post; and

posting the proposed post, with the suggested one or more revisions to the online social community.

17. The method of claim 16, wherein generating the prediction model comprises:

gathering a plurality of posts from one or more online social communities and at least one response to each post of the plurality of posts;

calculating a plurality of performance indicators for each post in the plurality of posts, wherein the plurality of performance indicators comprise a sentiment score for the at least one response and at least one of a number of responses to the post, a number of responders to the post, a longevity of the post, or a half-life of the post;

de-biasing the calculated plurality of performance indicators; and

building the prediction model using the de-biased plurality of performance indicators.

18. The method of claim 16, wherein suggesting, by the at least one processor, one or more revisions to the proposed post further comprises:

identifying one or more new terms based on the proposed post and the predicted community reaction;

predicting, by the at least one processor using the prediction model, a new sentiment score for each of the one or more new terms; and

suggesting, by the at least one processor, one or more revisions to the proposed post in light of the predicted sentiment score and the new sentiment score for each of the one or more new terms.

19. The method of claim 17, further comprising retraining the prediction model at predefined intervals of time.

20. The method of claim 17, wherein de-biasing comprises calculating a mean (m) and a standard deviation (s) of the performance indicators for responses to a given post and calculating a normalized score (r) to determine a de-biased score r′

, where r′

=(r−

m)/s.