Coreference Resolution in Machine Learning: A Survey

Coreference resolution (CR) is an important task in natural language processing (NLP), that aims to identify whether the different expressions from texts refer to a common entity. Specifically, Coreference resolution is an integral part of building better machines for text understanding and thus fostering impressive progress in machine comprehension, information retrieval, dialogue systems, sentiment analysis, document summarization, question answering, etc. This paper provides a comprehensive survey of CR techniques, which focus on rule-based systems, machine learning approaches, and recent advances in deep learning models. We discuss the challenges and advancements in CR for international, Indian, and regional languages and outline future research directions. In addition, we provide an overview of the benefits of existing research and the limitations that need to be addressed for further progress in this field.