Communication monitoring based on sentiment

Patent metadata

Field	Value
Publication number	US-9710459-B2
Application number	US-201514828553-A
Country	US
Kind code	B2
Filing date	Aug 18, 2015
Priority date	Aug 18, 2015
Publication date	Jul 18, 2017
Grant date	Jul 18, 2017

Abstract

Method and system are provided for communication monitoring based on sentiment. The method includes: monitoring a new communication prior to sending created by a sender to a recipient, wherein the new communication includes text input by the sender; and analyzing the text of the new communication using sentiment analysis and determining a sentiment score on a scale between negative sentiment and positive sentiment. In response to the sentiment score for the text of the new communication being on the negative side of a predefined threshold on the scale, referencing an overall relationship score based on past communications between the sender and the recipient. In response to the overall relationship score being on the negative side of a predefined threshold, holding a transmission of the new communication for further review.

First claim

What is claimed is:

1. A computer-implemented method for communication monitoring based on sentiment, comprising:
monitoring a new communication prior to sending created by a sender to a recipient, wherein the new communication comprises text input by the sender;
analyzing the text of the new communication using sentiment analysis and determining a sentiment score on a scale between negative sentiment and positive sentiment;
referencing an overall relationship score based on past communications between the sender and the recipient in response to the sentiment score for the text of the new communication being on a negative side of a first predefined threshold on the scale;
holding a transmission of the new communication or activating another action in response to the overall relationship score being on the negative side of a second predefined threshold; and
sending a notification to at least one of the sender or a supervisor of the sender in response to the overall relationship score being on the negative side of the second predefined threshold.

2. The method as claimed in claim 1 , wherein the overall relationship score is generated by performing sentiment analysis on text content of past communications between the sender and the recipient and determining the sentiment score on the scale between negative sentiment and positive sentiment.

3. The method as claimed in claim 1 , wherein the overall relationship score is generated by performing sentiment analysis on text content of past communication between the sender and the recipient from multiple communication channel sources.

4. The method as claimed in claim 1 , wherein the overall relationship score is based on a weighted average of sentiment analysis of text content from multiple communication channel sources.

5. The method as claimed in claim 1 , wherein multiple communication channel sources comprise sources from a group of: email communication channels, instant messaging communication channels, short message service communication channels, social media network communication channels, online forum channels, and other forms of communication channel sent from a computing device or mobile communication device.

6. The method as claimed in claim 1 , comprising updating the overall relationship score based on the sentiment score for the new communication.

7. The method as claimed in claim 1 , further comprising:
resubmitting an amended new communication for analyzing the text of the amended new communication using sentiment analysis and determining the sentiment score on the scale between negative sentiment and positive sentiment.

8. The method as claimed in claim 1 , further comprising:
selecting a signature for the communication from a plurality of signatures each having a different sentiment, wherein the selection is based on the sentiment score of the new communication.

9. A non-transitory computer program product for communication monitoring based on sentiment, the computer program product comprising:
one or more non-transitory computer-readable storage media and program instructions stored on the one or more non-transitory computer-readable storage media, the program instructions comprising:
program instructions to monitor a new communication prior to sending created by a sender to a recipient, wherein the new communication comprises text input by the sender;
program instructions to analyze the text of the new communication using sentiment analysis and program instructions to determine a sentiment score on a scale between negative sentiment and positive sentiment;
program instructions to reference an overall relationship score based on past communications between the sender and the recipient in response to the sentiment score for the text of the new communication being on a negative side of a first predefined threshold on the scale;
program instructions to hold a transmission of the new communication or activating another action in response to the overall relationship score being on the negative side of a second predefined threshold in response to the overall relationship score being on the negative side of a second predefined threshold; and
program instructions to send a notification to at least one of the sender or a supervisor of the sender in response to the overall relationship score being on the negative side of the second predefined threshold.

10. The computer program product as claimed in claim 9 , wherein the overall relationship score is generated by
program instructions to perform sentiment analysis on text content of past communications between the sender and the recipient and program instructions to determine the sentiment score on the scale between negative sentiment and positive sentiment.

11. The computer program product as claimed in claim 9 , wherein the overall relationship score is generated from
program instructions to perform sentiment analysis on text content of past communication between the sender and the recipient from multiple communication channel sources.

12. The computer program product as claimed in claim 9 , wherein the overall relationship score is based on
program instructions to create a weighted average of sentiment analysis of text content from multiple communication channel sources.

13. The computer program product as claimed in claim 9 , wherein multiple communication channel sources comprise sources from a group of: email communication channels, instant messaging communication channels, short message service communication channels, social media network communication channels, online forum channels, and other forms of communication channels sent from a computing device or mobile communication device.

14. The computer program product as claimed in claim 9 , comprising
program instructions to update the overall relationship score based on the sentiment score for the new communication.

15. The computer program product as claimed in claim 9 , further comprising:
program instructions to resubmit an amended new communication for analyzing the text of the amended new communication using sentiment analysis and program instructions to determine the sentiment score on the scale between negative sentiment and positive sentiment.

16. The computer program product as claimed in claim 9 , further comprising:
program instructions to select a signature for the communication from a plurality of signatures each having a different sentiment, wherein the program instructions to select is based on the sentiment score of the new communication.

17. A computer system for communication monitoring based on sentiment, the computer system comprising:
one or more computer processors, one or more non-transitory computer-readable storage media, and program instructions stored on one or more of the one or more non-transitory computer-readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to monitor a new communication prior to sending created by a sender to a recipient, wherein the new communication comprises text input by the sender;
program instructions to analyze the text of the new communication using sentiment analysis and program instructions to determine a sentiment score on a scale between negative sentiment and positive sentiment;
program instructions to reference an overall relationship score based on past communications between the sender and the recipient in response to the sentiment score for the text of the new communication being on a negative side of a first predefined threshold on the scale

Assignees

• IBM

Inventors

• Baldwin Timothy J
• Johnson Andrew
• Johnson Peter J
• Xu Fenglian

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F40/194

  Calculation of difference between files · CPC title

• H04L51/12

  Electricity · mapped topic

• G06F17/2785 Primary

  Physics · mapped topic