Abstract
News editors obtained the following quote from the background information suppli ed by the inventors: “Supervised classification task models are trained using ex amples consisting of utterances (strings of text in a natural language such as E nglish), and a label which classifies the utterance into one of several classes (commonly called intents in the literature). For example, the classification may be as simple as “Positive”, “Negative”, and “Neutral”, and one wishes to classi fy individual utterances based on the sentiment therein. “Yay that’s great” may be classified as “Positive” while “Terrible, just terrible” may be classified as “Negative”. This is one simple example, and in general the classification ontol ogy may be arbitrarily complex, with tens, hundreds, or more possible classes.
NETL
NSTL