SENTIMENT ANALYSIS

• US 20120191730A1
• Filed: 01/19/2012
• Published: 07/26/2012
• Est. Priority Date: 01/20/2011
• Status: Active Grant

• Current Assignee: IPC Systems Incorporated (Silver Lake Management Co. LLC)
• Original Assignee: IPC Systems Incorporated (Silver Lake Management Co. LLC)
• Inventors: Parikh, Mihir, Fabricant, Robert M., Hicks, Ed
• Granted Patent: US 9,208,502 B2
• Time in Patent Office: Days
• US Class Current: 707/754
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

Data is received from multiple data sources. At least one of the data sources is an active audio or video communication. The received data is analyzed by extracting instances of a keyword from the received data and analyzing contextual data near the keyword. Sentiment about the extracted keyword is gauged based on the contextual data. The derived sentiment data from the multiple data sources is aggregated, and an aggregated view of the derived sentiment data is presented.

Claims

1. A method for analyzing received data, comprising the steps of:

using at least one processor to perform;

receiving data from multiple data sources, wherein at least one of the data sources is an active audio or video communication;

analyzing the received data by extracting instances of a keyword from the received data and analyzing contextual data near the keywords;

gauging sentiment about the extracted keyword based on the contextual data;

aggregating the derived sentiment data from the multiple data sources; and

presenting an aggregated view to the user of the derived sentiment data.

2. The method according to claim 1, wherein the active audio or video communication is an active audio or video broadcast, and wherein voice-to-text transcription is used to extract keywords and contextual data from the active audio or video broadcast.

3. The method according to claim 2, wherein the text transcription of the audio or video broadcast is displayed along with the sentiment data.

4. The method according to claim 1, wherein a user can manually adjust sentiment for a keyword.

5. The method according to claim 1, wherein the keywords and contextual data are extracted from a live text-based communication.

6. The method according to claim 1, wherein positive and negative sentiment about the keyword are displayed as a graph of sentiment for the keyword.

7. The method according to claim 1, wherein a particular set of data sources can be selected to display sentiment data associated with those particular data sources.

8. The method according to claim 1, wherein one or more keywords are defined by a user, either directly or indirectly based on existing data sources in use by the user.

9. The method according to claim 1, wherein the aggregated view and derived sentiment data is shared among multiple users.

10. The method according to claim 1, wherein an alert is issued when a sentiment threshold for the keyword is reached.

11. A system for displaying a user interface, the system comprising at least one processor, wherein the processor is configured to:

receive data from multiple data sources, wherein at least one of the data sources is an active audio or video communication;

analyze the received data by extracting instances of a keyword from the received data and analyzing contextual data near the keywords;

gauge sentiment about the extracted keyword based on the contextual data;

aggregate the derived sentiment data from the data sources; and

present an aggregated view to the user of the derived sentiment data.

12. The system according to claim 11, wherein the active audio or video communication is an active audio or video broadcast, and wherein voice-to-text transcription is used to extract keywords and contextual data from the active audio or video broadcast.

13. The system according to claim 12, wherein the text transcription of the audio or video broadcast is displayed along with the sentiment data.

14. The system according to claim 11, wherein a user can manually adjust sentiment for a keyword.

15. The system according to claim 11, wherein the keywords and contextual data are extracted from a live text-based communication.

16. The system according to claim 11, wherein positive and negative sentiment about the keyword are displayed as a graph of sentiment for the keyword.

17. The system according to claim 11, wherein a particular set of data sources can be selected to display sentiment data associated with those particular data sources.

18. The system according to claim 11, wherein one or more keywords are defined by a user, either directly or indirectly based on existing data sources in use by the user.

19. The system according to claim 11, wherein the aggregated view and derived sentiment data is shared among multiple users.

20. The system according to claim 11, wherein an alert is issued when a sentiment threshold for the keyword is reached.

21. A non-transitory computer-readable medium having stored thereon sequences of instructions, the sequences of instructions including instructions, which, when executed by a processor, cause the processor to perform:

receiving data from multiple data sources, wherein at least one of the data sources is an active audio or video communication;

analyzing the received data by extracting instances of a keyword from the received data and analyzing contextual data near the keywords;

gauging sentiment about the extracted keyword based on the contextual data;

aggregating the derived sentiment data from the multiple data sources; and

presenting an aggregated view of the derived sentiment data.