About Keyword Extraction --

It can be said that other than the frequency, what makes a word a key is the specialty of the word to the document that the word exists in. If a
word appears much more frequently in a document than with respect to other documents, this can be
another distinguishing feature of that word on deciding whether it is a keyword or not.

Combining these two properties, we obtain the metric TFxIDF (standing for Term Frequency x Inverse
Document Frequency) score, which is the standard metric used in Information extraction [2], and for
a word W in document D, is defined as

TFxIDF (P, D) = P(word in D is W) x [ - log P(W in a document) ].

The first term in this formula is calculated by counting the number of times the word occurs in the
document and dividing it to the total number of words in it. The second term is calculated by counting
the number of documents in the training set that the word occurs in except D and dividing it by the
total number of documents in the training set.