A Comparative and Experimental Study on Automatic Question Answering Systems and its Robustness against Word Jumbling

We aim to understand and improve the performance of these state of the art Question answer generation using Natural Language models on different levels of noise. A model that Processing models is ubiquitous in the is robust to context corruption can increase the satisfaction world around us. It is used in many use cases such of the users of the system, especially for as the building of chat bots, suggestive prompts in people whose English is not their native language. It is and answer questions appropriately. This is highly relevant because a frequently asked questions relevant because many times we do not have control (FAQ) list can only have a finite amount of of the sources of data and how it is generated. In such cases, it becomes highly important to be able to answer new questions accurately as for the models to be able to understand and process long as it is a relevant question. In commercial this text which is riddled with noise.