syntactic parsing
In plain wordsWorking out the grammatical structure of a sentence — which words group together and which word modifies which — usually drawn as a tree.
The task of analyzing a sentence's grammatical structure and producing a parse tree. Constituency (phrase-structure) parsing groups words into nested phrases such as noun and verb phrases, in the tradition of context-free grammars; dependency parsing links each word to its syntactic head with labeled relations such as subject or object. Algorithms include chart parsers (CKY, Earley) with probabilistic grammars, graph-based dependency parsers that find maximum spanning trees, and fast transition-based parsers that build trees with shift–reduce actions, all later driven by neural scoring. Parsers are trained and evaluated on treebanks — corpora of hand-parsed sentences such as the Penn Treebank and the multilingual Universal Dependencies. Explicit parsing was once essential for translation, extraction and question answering; LLMslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. encode much syntax implicitly, and parsing now serves linguistics, structured-data pipelines and interpretability studies.
The dependency parse shows that 'with the telescope' attaches to 'saw', not to 'man'.
Origin
Noam Chomsky's 1950s work on formal grammars, together with later efficient context-free parsing algorithms by others (Cocke–Kasami–Younger in the 1960s; Jay Earley, 1970), underlies the constituency tradition, and Lucien Tesnière's dependency grammar (1959) the dependency one. Treebank-trained statistical parsers by Michael Collins and Eugene Charniak (late 1990s) and Joakim Nivre's transition-based parsing (2003) shaped modern practice.
Usage note
In computing generally, 'parsing' means analyzing any formal input, such as source code or JSON, against a grammar; LLM engineers also use it for extracting structured fields from model output.
Related
- part-of-speech taggingpart-of-speech tagging, n.: Labeling each word in a sentence with its grammatical role — noun, verb, adjective and so on — as in 'Time (noun) flies (verb) like (preposition) an arrow'.
- natural language processingnatural language processing, n.: The field that teaches computers to handle human language — to read, interpret, translate, summarize and produce text and speech. It is one of the oldest and largest areas of artificial intelligence.
- computational linguisticscomputational linguistics, n.: The science of human language studied with computers: using computational models to understand how language works, and building the language technology that grows out of that understanding.
- semantic parsingsemantic parsing, n.: Translating a sentence into a precise, machine-executable form — such as a database query or a logical formula — so that a computer can act on what was asked.
- corpuscorpus, n.: A large collection of written or spoken language, gathered so that people or computers can study it or learn from it — like a library assembled for a language model.
Sources
- Marcus, M. P., Santorini, B. & Marcinkiewicz, M. A. (1993). Building a Large Annotated Corpus of English: The Penn Treebank. Computational Linguistics 19(2), 313–330.
- Nivre, J. et al. (2016). Universal Dependencies v1: A Multilingual Treebank Collection. LREC 2016.
Mentioned in
natural language processingnatural language processing, n.: The field that teaches computers to handle human language — to read, interpret, translate, summarize and produce text and speech. It is one of the oldest and largest areas of artificial intelligence., computational linguisticscomputational linguistics, n.: The science of human language studied with computers: using computational models to understand how language works, and building the language technology that grows out of that understanding., Penn TreebankPenn Treebank, prop. n.: A collection of English text, best known for its million words of newspaper articles, in which experts marked each word's grammatical role and drew each sentence's structure as a tree; it trained and tested a generation of language software., part-of-speech taggingpart-of-speech tagging, n.: Labeling each word in a sentence with its grammatical role — noun, verb, adjective and so on — as in 'Time (noun) flies (verb) like (preposition) an arrow'., semantic parsingsemantic parsing, n.: Translating a sentence into a precise, machine-executable form — such as a database query or a logical formula — so that a computer can act on what was asked.