A Review on Approaches and Applications of Natural Language Inference
摘要
Natural Language Processing (NLP) is a wasp area in language processing, and Natural Language Inference (NLI) is a part of it. The idea of NLI is to find a relationship between two sentences. There are many applications built on NLP alike question answering, chat-bots, and many more. This review focuses on the fundamentals of NLI as well as some of its complications. It indicates exactly how traditional approaches have been replaced by modern approaches like deep learning and recurrent neural network. It also examines different components, limitations, and available datasets of NLI and reviews the progress of NLI in rapid terms. In summary, this paper surveys the evolution of NLI from beginning up till now with all the advancements and challenges.