N-grams are continuous sequences of words or symbols or tokens in a document. In technical terms, they can be defined as the neighboring sequences of items in a document. They come into play when we deal with text data in NLP(Natural Language Processing) tasks.
The term smoothing refers to the adjustment of the maximum likelihood estimator of a language model so that it will be more accurate. ... When estimating a language model based on a limited amount of text, such as a single document, smoothing of the maximum likelihood model is extremely important.
