At Expedia Group, we strive to improve our customer satisfaction by providing a frictionless shopping and support experience. A core part of this are machine learning models that power Expedia Group's virtual agents, that are available 24x7 to guide customers through any changes in their travel plans, pull-up information about their upcoming trips, book new trips and much more. Each capability has an associated'intent', which is identified by an intent classification model. However, natural language is messy, and not all customer utterances are actionable -- perhaps not even relevant. Therefore, the predicted probabilities of a classification model -- intent classifier in this case -- should be reliable: a misclassification should not have a high probability.

Customer service is rapidly becoming an essential focus for brands; Gartner research estimates that 89% of businesses will compete on customer experience rather than products or services. Furthermore, three in five Americans are willing to try a new company for a better customer relations experience. Your businesses' biggest assets, then, may very well be your customer relations and community management teams. One way to ensure your brand is competitive in its handling of customer service is its availability to consumers. Businesses typically hear from only 4% of dissatisfied customers--the rest remain silent, which means you're often without an opportunity to set things right and ensure they're happy before they switch to another business.