SYSTEM AND METHOD FOR SOCIAL BEHAVIOR MAPPING
A method for managing social media communications of an organization supported by a contact center system includes receiving a communication by a user from a social media platform, determining a relevance of the communication to the organization, analyzing the communication to generate a suggested response to the communication, routing the communication to an agent of the contact center system, and displaying the suggested response on a display device of the agent.
1 . A method for managing social media communications of an organization supported by a contact center system, the method comprising:
receiving, by a processor, a communication by a user from a social media platform;
determining, by the processor, a relevance of the communication to the organization;
analyzing, by the processor, the communication to generate a suggested response to the communication;
routing, by the processor, the communication to an agent of the contact center system; and
displaying, by the processor, the suggested response on a display device of the agent.
2 . The method of claim 1 , wherein determining the relevance of the communication comprises:
determining, by the processor, a score of the communication based on one or more criteria;
determining, by the processor, that the score is above a threshold; and
in response to determining that the score is above the threshold, queuing, by the processor, the communication for analysis and routing to an agent.
3 . The method of claim 2 , wherein the one or more criteria comprise one or more of sentiment, number of impressions, followers-to-following ratio, profile engagement, originality, and attentiveness.
4 . The method of claim 2 , further comprising:
in response to determining that the score is not above the threshold, discarding, by the processor, the communication without further involvement by any agent.
5 . The method of claim 1 , wherein the suggested response is predicted within a threshold level of confidence.
6 . The method of claim 1 , wherein analyzing the communication to generate the suggested response to the communication comprises:
maintaining, by the processor, a model for each section of a plurality of sections of a suggested response, each of the models modeling a correlation of a plurality of expression feature vectors to a plurality of candidate textual blocks;
identifying, by the processor, a particular section of the suggested response;
retrieving, by the processor, the model for the identified particular section;
providing, by the processor, an expression feature vector associated with the communication to the retrieved model;
receiving, by the processor, identification of one or more candidate textual blocks correlated to the input features vector; and
generating, by the processor, the suggested response based on the identification of one or more candidate textual blocks.
7 . The method of claim 6 , wherein the plurality of sections of the suggested response comprise an opening section, a main body section, and a closing section.
8 . The method of claim 6 , wherein the expression feature vector associated with the communication comprises one or more of a gender of the user, a geolocation of the communication, a time of the communication, a text of the communication, an originality of the communication, a sentiment score of the communication, a response timeliness between an original expression and a response to the expression, a known past activity of the user, a number of impressions of the user, and biographical data of the user.
9 . The method of claim 1 , further comprising:
displaying, by the processor, contextual information associated with the communication on the display device of the agent,
wherein the contextual information comprises information about an author of the communication and a general state of social expressions in aggregate.
10 . A system for managing social media communications of an organization supported by a contact center system, the system comprising:
a processor; and
a memory coupled to the processor, wherein the memory stores instructions that, when executed by the processor, cause the processor to perform:
receiving a communication by a user from a social media platform;
determining a relevance of the communication to the organization;
analyzing the communication to generate a suggested response to the communication;
routing the communication to an agent of the contact center system; and
displaying the suggested response on a display device of the agent.
11 . The system of claim 10 , wherein determining the relevance of the communication comprises:
determining a score of the communication based on one or more criteria;
determining that the score is above a threshold; and
in response to determining that the score is above the threshold, queuing the communication for analysis and routing to an agent.
12 . The system of claim 11 , wherein the one or more criteria comprise one or more of sentiment, number of impressions, followers-to-following ratio, profile engagement, originality, and attentiveness.
13 . The system of claim 11 , wherein the instructions further cause the processor to perform:
in response to determining that the score is not above the threshold, discarding the communication without further involvement by any agent.
14 . The system of claim 10 , wherein the suggested response is predicted within a threshold level of confidence.
15 . The system of claim 10 , wherein analyzing the communication to generate the suggested response to the communication comprises:
maintaining a model for each section of a plurality of sections of a suggested response, each of the models modeling a correlation of a plurality of expression feature vectors to a plurality of candidate textual blocks;
identifying a particular section of the suggested response;
retrieving the model for the identified particular section;
providing an expression feature vector associated with the communication to the retrieved model;
receiving identification of one or more candidate textual blocks correlated to the input features vector; and
generating the suggested response based on the identification of one or more candidate textual blocks.
16 . The system of claim 15 , wherein the plurality of sections of the suggested response comprise an opening section, a main body section, and a closing section.
17 . The system of claim 15 , wherein the expression feature vector associated with the communication comprises one or more of a gender of the user, a geolocation of the communication, a time of the communication, a text of the communication, an originality of the communication, a sentiment score of the communication, response timeliness between an original expression and a response to the expression, a known past activity of the user, a number of impressions of the user, and biographical data of the user.
18 . The system of claim 10 , wherein the instructions further cause the process to perform:
displaying contextual information associated with the communication on the display device of the agent,
wherein the contextual information comprises information about an author of the communication and a general state of social expressions in aggregate.
19 . A system for managing social media communications of an organization supported by a contact center system, the system comprising:
means for receiving a communication by a user from a social media platform;
means for determining a relevance of the communication to the organization;
means for analyzing the communication to generate a suggested response to the communication;
means for routing the communication to an agent of the contact center system; and
means for displaying the suggested response on a display device of the agent.