A Chatbot for a Dialogue-Based Second Language Learning System.

This paper presents a chatbot for a Dialogue-Based Computer-Assisted second Language Learning (DB-CALL) system. A DB-CALL system normally leads dialogues by asking questions according to given scenarios. User utterances outside the scenarios are normally considered as semantically improper and simply rejected. In this paper, we assume that raising the freedom of dialogue can stimulate the user's interest in learning. For this, a chatbot based on a search engine with a dialogue corpus has been developed to deal with conversations out of the scenarios. We evaluate the chatbot separately in two different cases: as an independent bot and as an auxiliary system. The results showed that, unlike the independent chatbot system, the chatbot as an auxiliary system showed a much lower turn success ratio.