Trigrams
Specific type of n-gram where n is 3, commonly used in language modeling and predicting the next item in NLP.
Trigrams are a crucial tool in Natural Language Processing (NLP), a sector of AI that deals with the interaction between computers and human language. In a trigram, a sequence of three items is taken from a given sample of text or speech. In the context of NLP, these 'items' are typically words, although they could also be letters, syllables, or other linguistic units, depending on the specific problem at hand. The prime usage of trigrams is for predicting the upcoming word in a sentence, which is crucial for language modeling, machine translation, spell check, and speech recognition.
Trigrams come under the umbrella concept of n-gram models which have been extensively used since the 1980s in fields like computational linguistics and probability theory.
Usage of trigrams as a language modeling tool was extensively adopted within the AI community due to the work of research groups focused on computational linguistics and NLP, such as Microsoft Research and Google's Natural Language Processing group.