Prediction of expected waiting time for business-agnostic contact centers using deep neural networks

摘要

The method (800) for predicting the estimated waiting time (130) comprises the step of receiving a pending support request (120) from the user (10). A pending support request is associated with a number of high-level features (202) with the number of active support agents (202a), the number of available support agents (202b), and the queue depth (202c). .. The method also comprises a step of predicting an estimated wait time (130) for a user with a pending support request using a wait time prediction model (270) that receives a plurality of high-level features as feature inputs. The latency prediction model is trained on a corpus of learning support requests (120H) with corresponding high-level features and corresponding actual latency (202g). The method further comprises a step of providing the user with an estimated waiting time indicating an estimated duration before responding to a pending support request.