“NLTK: The Natural Language Toolkit” was presented by Steven Bird and Edward Loper at the 42nd Annual Meeting of the Association for Computational Linguistics (ACL) in Barcelona in July 2004, in the interactive poster and demonstration sessions (pages 214-217). An earlier description had appeared at a 2002 ACL teaching workshop.
NLTK is an open-source suite of Python libraries and programs for symbolic and statistical natural language processing. It bundles tokenizers, taggers, parsers, classifiers, and a large set of corpora and lexical resources, including an interface to WordNet. Bird and Loper built it first as a teaching tool: the toolkit’s clear, well-documented modules let students see each step of an NLP pipeline rather than treating it as a black box.
NLTK became the standard entry point for learning natural language processing in Python and underpinned the widely used book “Natural Language Processing with Python.” Production systems later shifted to faster libraries such as spaCy and, eventually, transformer models, but NLTK remained the teaching standard and a convenient toolkit for linguistic preprocessing and experimentation.