A Review on Conversational Question Answering (CQA)

The internet is constantly changing how people communicate with one another, manage their daily lives, and share the information. As a result of digitization, automated Question Answering Systems are becoming more important in extracting useful information from the knowledge sources. In general, Question Answering (QA)mechanisms are intended to answer a specific question only once (so-called single-turn).However, to satisfy a user's information needs, the Conversational Question Answering (CQA) system must grasp the given context and engage in multi-turn QA. While single-turn QA leads the majority of current research, multi-turn QA has rapidly gained prominence due to the availability of large-scale multi-turn QA datasets and the development of “pre-trained language models”. In this research paper, we have presented a comprehensive review on the Conversational Question Answer (CQA) field which includes need of CQA followed by categorization of QA system, list of available datasets and various evaluation metrics.