Dynamic prioritization of collaboration between human and virtual agents
A method of dynamic prioritization of collaboration between human and virtual agents according to an embodiment includes generating a priority score for a conversation between a human user and a chat bot, wherein the priority score is based on at least a sentiment of the human user and a temporal aspect of the conversation, providing the priority score to a human agent via an interface for monitoring the priority score, transferring control of the conversation from the chat bot to the human agent based on an interaction of the human agent with the interface, and yielding control of the conversation from the human agent back to the chat bot in response to another interaction of the human agent with the interface.
1. A method of dynamic prioritization of collaboration between human and virtual agents, the method comprising:
generating a priority score for a conversation between a human user and a chat bot, wherein the priority score is based on at least a first metric associated with a real-time sentiment of the human user and a second metric associated with a real-time response wait time of the human user in the conversation, wherein the second metric is calculated as a message rate of the conversation multiplied by a current wait time of the human user in the conversation;
providing the priority score to a human agent via an interface for monitoring the priority score;
transferring control of the conversation from the chat bot to the human agent based on an interaction of the human agent with the interface; and
yielding control of the conversation from the human agent back to the chat bot in response to another interaction of the human agent with the interface.
2. The method of claim 1 , wherein the priority score is a weighted sum of at least the first metric and the second metric.
3. The method of claim 2 , wherein the first metric is a function of the real-time sentiment of the human user, wherein the real-time sentiment of the human user has an output value between a negative scalar indicative of a maximum negative sentiment and a positive scalar indicative of a maximum positive sentiment; and
wherein the first metric is calculated according to f sen (C)=max{0, −sen(C)}, wherein f sen (C) is the first metric, sen(C) is the real-time sentiment of the human user, and C is the conversation.
4. The method of claim 2 , wherein the first metric is a function of the real-time sentiment of the human user, wherein the real-time sentiment of the human user has an output value between a negative scalar indicative of a maximum negative sentiment and a positive scalar indicative of a maximum positive sentiment; and
wherein the first metric is calculated according to f sen (C)=max{0, sen(C)}, wherein f sen (C) is the first metric, sen(C) is the real-time sentiment of the human user, and C is the conversation.
5. The method of claim 2 , wherein the first metric is a function of the real-time sentiment of the human user, wherein the real-time sentiment of the human user has an output value between a negative scalar indicative of a maximum negative sentiment and a positive scalar indicative of a maximum positive sentiment; and
wherein the first metric is calculated according to f sen (C)=|sen(C)|, wherein f sen (C) is the first metric, sen(C) is the real-time sentiment of the human user, and C is the conversation.
6. The method of claim 1 , wherein the priority score is further based on a topic of the conversation; and
wherein the priority score is a weighted sum of the first metric, the second metric, and a third metric associated with the topic of the conversation.
7. The method of claim 1 , further comprising performing bot-driven interaction pacing to the conversation by providing filler content by the chat bot based on the priority score.
8. The method of claim 1 , wherein generating the priority score for the conversation comprises generating the priority score based on a sequence of time-stamped messages that constitute the conversation.
9. The method of claim 1 , wherein the priority score is a based on a plurality of weighted metrics; and
wherein one or more weights of the plurality of weighted metrics is adjustable by the human agent.
10. A system for dynamic prioritization of collaboration between human and virtual agents, the system comprising:
at least one processor; and
at least one memory comprising a plurality of instructions stored thereon that, in response to execution by the at least one processor, causes the system to:
generate a priority score for a conversation between a human user and a chat bot, wherein the priority score is based on at least a first metric associated with a real-time sentiment of the human user and a second metric associated with a real-time response wait time of the human user in the conversation, wherein the second metric is calculated as a message rate of the conversation multiplied by a current wait time of the human user in the conversation;
provide the priority score to a human agent via an interface for monitoring the priority score;
transfer control of the conversation from the chat bot to the human agent based on an interaction of the human agent with the interface; and
yield control of the conversation from the human agent back to the chat bot in response to another interaction of the human agent with the interface.
11. The system of claim 10 , wherein the priority score is a weighted sum of at least the first metric and the second metric.
12. The system of claim 11 , wherein the first metric numerically prioritizes one of negative real-time sentiment of the human user, positive real-time sentiment of the human user, or extreme real-time sentiment of the human user.
13. The system of claim 10 , wherein the priority score is further based on a topic of the conversation; and
wherein the priority score is a weighted sum of the first metric, the second metric, and a third metric associated with the topic of the conversation.
14. The system of claim 10 , wherein the plurality of instructions further causes the system to perform bot-driven interaction pacing to the conversation by providing filler content by the chat bot based on the priority score.
15. The system of claim 10 , wherein the plurality of instructions further causes the system to provide a plurality of priority scores for corresponding conversations between human users and chat bots to the human agent via the interface for monitoring the priority scores.
16. The system of claim 10 , wherein the priority score is a based on a plurality of weighted metrics; and
wherein one or more weights of the plurality of weighted metrics is adjustable by the human agent.
17. A method of dynamic prioritization of collaboration between human and virtual agents, the method comprising:
generating a priority score for a conversation between a human user and a chat bot as a weighted sum of at least a first metric associated with a real-time sentiment of the human user, a second metric associated with a real-time response wait time of the human user in the conversation, wherein the second metric is calculated as a message rate of the conversation multiplied by a current wait time of the human user in the conversation, and a third metric associated with a topic of the conversation;
providing the priority score to a human agent via an interface for monitoring the priority score;
transferring control of the conversation from the chat bot to the human agent based on an interaction of the human agent with the interface; and
yielding control of the conversation from the human agent back to the chat bot in response to another interaction of the human agent with the interface.
18. The method of claim 1 , wherein the message rate is a number of messages of the conversation per unit of time.