System and method for an optimized, self-learning and self-organizing contact center
A system and method for an optimized, self-learning and self-organizing contact center has been developed. This system and method uses principles and tools of information theory, including the latent Dirichlet allocation which reduces information to specific predetermined topics and a distribution of topic related words to infer its hidden, generative underpinnings so to self-organize a contact center, infer its desired electronic versus human make up, and optimally route all customer requests to an electronic resource or a specific human agent best suited to respond to the request for maximal business value per interaction.
1. A self-learning and self-organizing contact center routing system comprising:
a topic-based routing module stored in a memory of and operating on a processor of a computing device; and
an interaction information optimization module stored in a memory of and operating on a processor of a computing device;
wherein the interaction information optimization module:
receives a plurality of incoming communications to a contact center, and for each incoming communication:
analyzes the incoming communication using a probabilistic model to identify a topic within the incoming communication;
sends the incoming communication and the topic to the topic-based routing module;
receives an outgoing communication from the topic-based routing module sent in response to the incoming communication;
determines an effectiveness of the outgoing communication by comparing the outgoing communication with a plurality of similar outgoing communications on the topic; and
updates a ranking of a plurality of human agents in the contact center based on each agent's proficiency with the topic; and
wherein the topic-based routing module:
receives the incoming communication;
routes the incoming communication to one of the plurality of human agents based on the topic and the ranking;
receives an outgoing communication, the outgoing communication comprising a response to the incoming communication from the human agent to whom the incoming message was routed; and
sends a copy of the outgoing communication to the interaction information optimization module.
2. The system of claim 1 , wherein the probabilistic model is a latent Dirichlet allocation.
3. The system of claim 1 , wherein the effectiveness of the outgoing communication is determined by comparisons with a plurality of similar outgoing communications based on one or more of the following: an information gap between information requested and information given in the outgoing communication, an end result of an incoming interaction, and efficiency of resource usage.
4. The system of claim 1 , wherein the proficiency of each agent is determined based on information contained in one or more of the following: the agent's curriculum vitae, training certificates, in-house work projects, and prior interactions of the agent with customers.
5. A method for self-learning and self-optimizing contact center routing, comprising the steps of:
creating a ranking of a plurality of human agents in the contact center based on each agent's proficiency with one or more topics; and
at an interaction information optimization module stored in a memory of and operating on a processor of a computing device:
intercepting a plurality of incoming communications to a contact center, and for each intercepted communication:
analyzing the intercepted communication using a probabilistic model to identify a topic within the incoming communication;
sending the incoming communication and the topic to a topic-based routing module stored in a memory of and operating on a processor of a computing device;
receiving an outgoing communication from the topic-based routing module sent in response to the incoming communication;
determining an effectiveness of the outgoing communication by comparing the outgoing communication with a plurality of similar outgoing communications on the topic; and
updating the ranking based on the effectiveness; and
at the topic-based routing module:
receiving the incoming communication;
routing the incoming communication to one of the plurality of human agents based on the topic and the ranking;
receiving an outgoing communication, the outgoing communication comprising a response to the incoming communication from the human agent to whom the incoming message was routed; and
sending a copy of the outgoing communication to the interaction information optimization module.
6. The method of claim 5 , wherein the probabilistic model is a latent Dirichlet allocation.
7. The method of claim 5 , wherein the effectiveness of the outgoing communication is determined by comparisons with a plurality of similar outgoing communications based on one or more of the following: an information gap between information requested and information given in the outgoing communication, an end result of an incoming interaction, and efficiency of resource usage.
8. The method of claim 5 , wherein the proficiency of each agent is determined based on information contained in one or more of the following: the agent's curriculum vitae, training certificates, in-house work projects, and prior interactions of the agent with customers.