A Review of Dialogue Intent Identification Methods for Closed Domain Conversational Agents

Abstract

Dialogue intent identification is an indispensable part of every conversational or a dialogue system. Intent identification is the process of deducing the goal or meaning of the sentence. Intent identification is performed using various classification algorithms. Performance of dialogue systems is vastly dependent on the accuracy of these intent identification methods and algorithms. Thus we review some of the available dialogue intent identification methods, train the classification models on a common dataset and then evaluate on the basis of various performance metrics. A comprehensive comparative study of various intent identification methods is obtained.