Natural Language Processing
Natural language processing (NLP) is the field of using computers to analyze, generate or transform human language. It includes tasks such as classification, extraction, translation, summarization and dialogue, using rules as well as learned models.
Also known as: NLP
A support workflow can detect a message’s language, extract a product name, classify the issue and draft a response. Each stage is a language-processing task, and each can fail differently. The same workflow can combine hand-written rules with statistical or neural models.
For example, a document system may identify personal names before sending text for human redaction. Finding every sensitive span matters more than producing a fluent explanation. A generative model can help with this task, but its prose quality does not measure extraction accuracy.
A large language model is one tool used within NLP, while a chatbot is an application that presents a conversational interface. Neither defines the entire field. This matters when choosing a solution: a narrow classifier or extraction pipeline may meet the task without a general assistant. Evaluate the required language behavior directly, including omissions and errors that a readable response can hide.
Sources
- Jurafsky and Martin: Speech and Language Processing — Covers language modeling, classification, retrieval, translation, extraction and dialogue across multiple methods.
Go deeper
- spaCy: Linguistic features docs
Inspect a practical pipeline for tokens, dependencies, entities and text representations.