We took a detailed look at part of speech tagging in part 1 and chunking in part 2 of this tutorial series. In this tutorial, we will learn about what parsing is, its different types, and techniques to automatically infer the parse tree of sentences.

Deep Parsing

Natural language parsing (also known as deep parsing) is a process of analyzing the complete syntactic structure of a sentence. This includes how different words in a sentence are related to each other, for example, which words are the subject or object of a verb. Probabilistic parsing uses language understanding such as grammatical rules. Alternatively, it may also use supervised training set of hand-parsed sentences to try to infer the most likely syntax and structure of new sentences.

Parsing is used to solve various complex NLP problems such as conversational dialogues and text summarization. It is different from 'shallow parsing' in that it yields more expressive structural representations which directly capture long-distance dependencies and underlying predicate-argument structures.

There are two main types of parse tree structures - constituency parsing and dependency parsing.