System and method for determining sentiment from text content

• US 8,838,438 B2
• Filed: 04/29/2011
• Issued: 09/16/2014
• Est. Priority Date: 04/29/2011
• Status: Active Grant

• Current Assignee: CBS Interactive Inc. (Paramount Global (f/k/a ViacomCBS Inc.))
• Original Assignee: CBS Interactive Inc. (Paramount Global (f/k/a ViacomCBS Inc.))
• Inventors: Leary, Adam, Pennie, Philip
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US13/098,302
• Publication Number:
  US 20120278064A1
• Time in Patent Office: 1,236 Days
• Field of Search: None
• US Class Current: 704/9
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique

Abstract

A system and method for determining sentiment from user-generated text content is provided. A sentiment score is determined for one or more terms in a user-generated text content. A sentiment value is determined for the text content that is based at least in part on the sentiment score for the one or more terms.

Claims

1. A method for determining sentiment from user-generated text content, the method being performed by one or more processors and comprising:

using the one or more processors to perform steps comprising;

determining a subject of a user-generated text content including one or more terms;

determining a sentiment value for the user-generated text content, wherein determining the sentiment value includes (a) determining a sentiment score for each of the one or more terms in the user-generated text content, and (b) weighing the sentiment score for each of the one or more terms based on a relationship between the term and the subject and a proximity of the term to the subject; and

associating the sentiment value with the subject.

2. The method of claim 1, wherein the subject corresponds to a business establishment.

3. The method of claim 1, wherein the sentiment value is associated with a particular sentiment category.

4. The method of claim 1, wherein determining the sentiment value includes determining a category-specific sentiment value for each of multiple sentiment categories.

5. The method of claim 1, wherein the sentiment value is mapped to one or more of multiple types of sentiments.

6. The method of claim 5, wherein the multiple types of sentiments include two or more of disgust, sadness, anger, fear or happiness.

7. The method of claim 6, wherein the sentiment value is further mapped to an intensity of a corresponding type of sentiment in the multiple types of sentiments.

8. The method of claim 1, wherein the sentiment value correlates to a single sentiment.

9. The method of claim 1, wherein the sentiment score for each of the one or more terms is predetermined.

10. The method of claim 1, further comprising presenting content that is based on the user-generated text content, the content including a feature that is informative of the sentiment value that is determined for the user-generated text content.

11. The method of claim 1, wherein the user-generated text content corresponds to a user feedback of a business establishment or product, and wherein the sentiment value corresponds to a sentiment of a user'"'"'s experience with the business establishment or product.

12. The method of claim 11, wherein the sentiment value indicates whether the user'"'"'s experience with the business establishment or product was positive or negative.

13. The method of claim 11, wherein the sentiment value indicates an intensity of the sentiment of the user'"'"'s experience with the business establishment or product.

14. The method of claim 11, wherein the business establishment or product corresponds to a restaurant or food item.

15. A system for determining sentiment from user-generated text content, the system comprising:

one or more processors; and

one or more memory resources storing instructions that, when executed by the one or more processors, cause the one or more processors to;

analyze a user-generated text content, including determining (i) a subject of the user-generated text content, (ii) a sentiment score for individual terms that are present in the user-generated text content, and (iii) a sentiment value for the user-generated text content based at least in part on the determined sentiment score for the individual terms, wherein determining (iii) includes weighing the sentiment score for each of the individual terms based on a relationship between the term and the subject and a proximity of the term to the subject; and

associate the sentiment value with the subject.

16. The system of claim 15, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to store data that is based on the sentiment value in association with the determined subject.

17. The system of claim 15, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to store the sentiment score for a list of terms.

18. The system of claim 15, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to identify one or more terms that are salient to a domain of the subject for text content that is user-generated.

19. The system of claim 15, wherein the subject corresponds to a business establishment or product.

20. The system of claim 15, wherein the subject corresponds to a restaurant or food item.

21. The system of claim 15, wherein the sentiment score indicates an intensity for a user'"'"'s like or dislike for the subject of the user-generated text content.

22. A method for determining sentiment from user-generated text content, the method being performed by one or more processors and comprising:

using the one or more processors to perform steps comprising;

determining a subject in a user-generated text content;

determining one or more sentiment terms in the user-generated text content;

determining a relevancy of each of the one or more sentiment terms to the subject;

determining a sentiment score for each of the one or more sentiment terms; and

determining a sentiment value for the subject based on one or more of the sentiment scores for each of the one or more sentiment terms that are determined to be relevant to the subject, wherein each of the sentiment scores are weighed based on a relationship between the sentiment term and the subject and a proximity of the sentiment term to the subject.

23. A non-transitory computer-readable medium that stores instructions for determining sentiment from user-generated text content, the instructions being executable by one or more processors for performing steps comprising:

determining a subject of a user-generated text content;

determining a sentiment value for the user-generated text content, wherein determining the sentiment value includes (a) determining a sentiment score for one or more terms in the user-generated text content, and (b) weighing the sentiment score for each of the one or more terms based on a relationship between the term and the subject and a proximity of the term to the subject; and

associating the sentiment value with the subject.