Improve Text Classification Accuracy with Intent Information

In addition, existing text classification approaches only consider the utterances in the coarse granularity level, which In recent years, goal-oriented dialogue systems have been may less the possibility of the model to explore relationship widely applied in intelligent voice assistant, e.g., Apple Siri, between token-level information and label information. For Amazon Alexa, where intent classification technology plays a example, in Figure 1, if there is a token "When" in the crucial part. Given input utterance in natural language, the intent input sentence, then the "NUM" intent is more likely to be classification module aims to detect the user's intent [10], recognized as high correlation with it, while the "LOC" intent [11], [14], [23]. Previous works have been proposed for better does not.