Natural Language Processing A Study of State of the Art
Nipun Sharma, Swati Sharma
Abstract
The world today is data driven. Data are increasing exponentially at lightning speed. The ever-increasing data are mostly unstructured. Based on data analytics, 80% of any enterprise data are unstructured, mostly human generated in the form of images, videos, speech and text that are natural language too. In addition, these data are first-hand, unbiased, uninfluenced and present the raw and authentic sentiments of the group that generated these data. In the current year, the number of smartphone users worldwide is 6.92 billion, which translates to 86.29% of the world’s population. This leads us to believe that this data form has immense potential and can lead to very important and ground-breaking information, which is otherwise very difficult to source. With this notion, the processing of natural data is a domain that demands a critical and meticulous understanding of the concepts and possibilities that revolve around it. Considering these facts, we have attempted to consolidate all the key concepts, rules and state of the art in the domain of natural language processing for better and improved human–computer interaction. This chapter presents an extensive presentation, demonstration and understanding of the current natural language processing scenario with the flavor of artificial intelligence, machine learning and current state of the art.