Dealing with Trouble: A Data-Driven Model of a Repair Type for a Conversational Agent

SLA, I propose a data-driven approach inspired by Conversation Analysis (CA) to create models of linguistic repair. Conversational agents for educational purposes, specifically I use the data set of instant messaging dialogues in German for Second Language Acquisition (SLA) use different approaches described in (Danilava et al. 2013). The corpus consists to support language learning through conversation. of 72 free conversations produced by 9 learners and CSIEC chatbot (Jia 2009) can correct spelling errors.