Which of the following best defines the TF-IDF approach?
An algorithm used for document classification based on word frequency.
A method for evaluating word importance based on their frequency within a document.
A technique for clustering text data using advanced neural networks.
A statistical measure calculating the correlation between words in a sentence.