Insurance Question Answering via Single-turn Dialogue Modeling

Abstract

With great success in single-turn question answering (QA), conversational QA is currentlyreceiving considerable attention. Several studies have been conducted on this topic fromdifferent perspectives. However, building areal-world conversational system remains achallenge. This study introduces our ongoingproject, which uses Korean QA data to develop a dialogue system in the insurance domain. The goal is to construct a system thatprovides informative responses to general insurance questions. We present the current results of single-turn QA. A unique aspect ofour approach is that we borrow the conceptsof intent detection and slot filling from taskoriented dialogue systems. We present detailsof the data construction process and the experimental results on both learning tasks.