DATA ANALYTICS SYSTEMS AND METHODS WITH PERSONALIZED SENTIMENT MODELS

• US 20170243120A1
• Filed: 02/23/2016
• Published: 08/24/2017
• Est. Priority Date: 02/23/2016
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Doan, Dai Duong, Merritt, Alexander
• Granted Patent: US 10,713,588 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A system for processing social media data includes a platform with a social media acquisition module configurable to collect a plurality of social media statements. The platform further includes an analysis engine configurable to analyze the plurality of social media statements according to a first sentiment model to generate first analytics data. The analysis engine is configurable to present the first analytics data to a client user, including a display of a sentiment value for each of the social media statements. The platform further includes a feedback queue configurable to receive feedback from the client user on at least a portion of the sentiment values; a model modification module configurable to modify the first sentiment model based on the feedback to result in a modified sentiment model; and a database configurable to store the modified sentiment model as a personalized sentiment model for the client user.

Claims

1. A system for processing social media data, comprising:

a computer, having a processor and a computer readable storage medium storing computer readable instructions capable of, upon execution by the processor, forming a platform comprising;

a social media acquisition module configurable to collect a plurality of social media statements;

an analysis engine configurable to analyze the plurality of social media statements according to a first sentiment model to generate first analytics data, the analysis engine configurable to present the first analytics data to a client user, including a display of a sentiment value for each of the social media statements;

a feedback queue configurable to receive feedback from the client user on at least a portion of the sentiment values;

a model modification module configurable to modify the first sentiment model based on the feedback to result in a modified sentiment model; and

a database configurable to store the modified sentiment model as a personalized sentiment model for the client user.

2. The system of claim 1,

wherein analysis engine is configurable to analyze the plurality of social media statements according to the personalized sentiment model to generate second analytics data and to present the second analytics data to the client user, including a display of the sentiment value for each of the social media statements,

the feedback queue configurable to receive further feedback from the client user on at least a further portion of the sentiment values,

the model modification model configurable to modify the second sentiment model based on the further feedback to result in a further modified sentiment model, and

a database configurable to store the further modified sentiment model as the personalized sentiment model for the client user.

3. The system of claim 1, wherein the analysis engine is configurable to present the first analytics data as a report listing the sentiment value for each of the social media statements.

4. The system of claim 3, wherein the analysis engine is configurable to present, for each of the social media statements, a statement tree formed by a plurality of nodes representing elements of the respective social media statement.

5. The system of claim 4, wherein the analysis engine is configurable to present, for each statement tree, a respective sentiment value associated with each of the plurality of nodes.

6. The system of claim 5, wherein the model modification module is configurable to modify at least one of the sentiment values associated with the plurality of nodes.

7. The system of claim 6, wherein the feedback queue is configurable to receive the feedback in the form of a feature vector for each of the plurality of nodes that has been modified.

8. The system of claim 7, wherein the analysis engine comprises a parser configurable to parse each of the social media statements and a classification module configurable to classify each of the elements with the respective sentiment value.

9. The system of claim 1, wherein the analysis engine is configurable to apply the first sentiment model as a recursive neural network.

10. The system of claim 1, wherein the first sentiment model is a basic sentiment model received from an analysis provider.

11. A computer implemented method for providing a personalized sentiment model for online social network analysis, the method comprising:

collecting a plurality of social media statements;

analyzing the plurality of social media statements according to a first sentiment model to generate first analytics data;

presenting, on a computing device, the first analytics data to a client user, including a display of a sentiment value for each of the social media statements;

receiving, via the computing device, feedback from the client user on at least a portion of the sentiment values;

modifying the first sentiment model based on the feedback to result in a modified sentiment model; and

storing the modified sentiment model as a personalized sentiment model for the client user.

12. The computer implemented method of claim 11, further comprising:

analyzing the plurality of social media statements according to the personalized sentiment model to generate second analytics data;

presenting, on the computing device, the second analytics data to the client user, including a display of the sentiment value for each of the social media statements;

receiving, via the computing device, further feedback from the client user on at least a further portion of the sentiment values;

modifying the second sentiment model based on the further feedback to result in a further modified sentiment model; and

storing the further modified sentiment model as the personalized sentiment model for the client user.

13. The computer implemented method of claim 11, wherein the presenting step includes presenting the first analytics data as a report listing the sentiment value for each of the social media statements.

14. The computer implemented method of claim 13, wherein the presenting step further includes displaying, for each of the social media statements, a statement tree formed by a plurality of nodes representing elements of the respective social media statement.

15. The computer implemented method of claim 14, wherein the presenting step further includes displaying, for each statement tree, a respective sentiment value associated with each of the plurality of nodes.

16. The computer implemented method of claim 15, wherein the modifying step includes modifying at least one of the sentiment values associated with the plurality of nodes.

17. The computer implemented method of claim 16, wherein the receiving step includes receiving the feedback in the form of a feature vector for each of the plurality of nodes that has been modified.

18. The computer implemented method of claim 17, wherein the analyzing step includes parsing each of the social media statements and classifying each of the elements with the respective sentiment value.

19. The computer implemented method of claim 11, wherein the analyzing step includes applying the first sentiment model as a recursive neural network.

20. The computer implemented method of claim 11, further comprising receiving the first sentiment model as a basic sentiment model from an analysis provider.