Sorting and displaying documents according to sentiment level in an online community

• US 10,565,401 B2
• Filed: 11/25/2014
• Issued: 02/18/2020
• Est. Priority Date: 02/28/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Berube, Richard E., Loredo, Robert E., Lu, Fang, Ricercato, Susan, Sardone, Zi
• Primary Examiner(s): Paula, Cesar B
• Assistant Examiner(s): Barnes, Jr., Carl E
• Application Number:
  US14/552,486
• Publication Number:
  US 20150248424A1
• Time in Patent Office: 1,911 Days
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06F 40/103 Formatting, i.e. changing of presentation of documents automatic justification G06F40/189; automatic line break hyphenation G06F40/191
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

An approach is described for sorting and displaying documents according to sentiment level in an online community. An associated method may include selecting a review topic in an online community and identifying a plurality of documents contributed for the review topic. The plurality of documents may include at least one of a product review submission, a marketing survey submission, a social network activity stream post, a discussion forum post, a weblog post, and an audiovisual sample. The method further may include obtaining sentiment data associated with each of the plurality of documents developing a sentiment model based on the obtained sentiment data. Additionally, the method may include organizing and presenting the plurality of documents in an online community interface based on the sentiment model.

Claims

1. A method comprising:

identifying a plurality of documents contributed for a review topic selected in an online community;

obtaining sentiment data associated with each of the plurality of documents, wherein obtaining the sentiment data comprises selecting content associated with each of the plurality of documents and determining n-grams in each of the plurality of documents from the selected content via a computing application capable of natural language processing;

developing a sentiment model based on the obtained sentiment data, wherein developing the sentiment model comprises, for each document among the plurality of documents, calculating a sentiment score for each n-gram in the document by establishing predefined rules based upon factors selected at least in part from the group consisting of (i) level of bias, (ii) use of figures of speech, and (iii) use of indicia of emphasis and by assigning a sentiment score to the n-gram consequent to comparing the n-gram relative to other n-grams in the context of considering the one or more factors, and calculating a sentiment score for the document based upon the respective sentiment scores calculated for the n-grams of the document; and

based on the sentiment model, organizing and presenting the plurality of documents in an online community interface of a client computing application, wherein organizing and presenting the plurality of documents comprises sorting the plurality of documents by assigning respectively higher priority values to documents that have respectively lower sentiment scores calculated during the development of the sentiment model.

2. The method of claim 1, wherein obtaining the sentiment data associated with each of the plurality of documents further comprises:

sending, to a sentiment processing application, the selected content associated with each of the plurality of documents; and

receiving, from the sentiment processing application, sentiment data corresponding to the n-grams in each of the plurality of documents.

3. The method of claim 2, wherein obtaining the sentiment data associated with each of the plurality of documents further comprises:

creating sentiment context data for each of the plurality of documents based on the received sentiment data; and

appending the sentiment context data created for each of the plurality of documents to associated metadata.

4. The method of claim 1, wherein calculating the sentiment score for the document based upon the respective sentiment scores calculated for the n-grams of the document comprises determining a sum of the respective sentiment scores calculated for all of the n-grams of the document.

5. The method of claim 1, wherein organizing and presenting the plurality of documents based on the sentiment model further comprises:

differentiating priority values of any group of documents assigned a same priority value by applying at least one tiebreak technique.

6. The method of claim 1, wherein the plurality of documents include at least one of (i) a product review submission, (ii) a marketing survey submission, (iii) a social network activity stream post, (iv) a discussion forum post, (v) a weblog post, and (vi) an audiovisual sample.

7. The method of claim 1, wherein a client associated with the client computing application is provided an option to determine relative weights of the one or more factors.

8. The method of claim 1, wherein a client associated with the client computing application is provided an option to determine how the one or more factors are considered in establishing the predefined rules.

9. The method of claim 1, wherein a client associated with the client computing application is provided an option to adjust which of the one or more factors are considered in establishing the predefined rules.

10. The method of claim 1, wherein calculating the sentiment score for the document based upon the respective sentiment scores calculated for the n-grams of the document comprises determining an average of the respective sentiment scores calculated for all of the n-grams of the document.

11. The method of claim 1, wherein organizing and presenting the plurality of documents based on the sentiment model further comprises:

displaying the sorted plurality of documents in the online community interface, wherein documents assigned respectively higher priority values are displayed more prominently than documents assigned respectively lower priority values.

12. The method of claim 11, wherein any document among the plurality of documents assigned a priority value lower than a designated threshold value is omitted from display.

13. The method of claim 11, wherein any document among the plurality of documents having a sentiment score higher than a designated threshold value is omitted from display.