The Trajectory of Natural Language Processing
Rachel Wagner-Kaiser, Tim Cerino · 2025
In this chapter, an overview is provided of the history of natural language processing (NLP). The history is broken down into three “eras”: classic NLP, modern NLP, and generative NLP. It covers the major advances from regular expression and rule-driven NLP, statistical and machine learning techniques, the more recent advances of deep learning applied at scale, and the emergence of large language models (LLMs). This chapter also covers best practices to select the most appropriate tool for a given purpose or situation, encouraging readers to think critically about the intersection of business need, data, and tools.