Interactive Classification by Asking Informative Questions
This addresses intent classification for users by enabling interactive refinement, but it is incremental as it builds on existing classification methods with added interaction.
The paper tackles interactive natural language classification by having a system ask informative binary or multi-choice questions after an initial user query, balancing between asking more questions and making predictions. It shows benefits of interaction and learning this balance, evaluated on two domains with unspecified numerical results.
We study the potential for interaction in natural language classification. We add a limited form of interaction for intent classification, where users provide an initial query using natural language, and the system asks for additional information using binary or multi-choice questions. At each turn, our system decides between asking the most informative question or making the final classification prediction.The simplicity of the model allows for bootstrapping of the system without interaction data, instead relying on simple crowdsourcing tasks. We evaluate our approach on two domains, showing the benefit of interaction and the advantage of learning to balance between asking additional questions and making the final prediction.