Why did you use that IDF formula?

whoosh-community / whoosh

Whoosh is a fast, featureful full-text indexing and searching library implemented in pure Python.

Other

244 stars 37 forks source link

Open fortable1999 opened 6 years ago

fortable1999 commented 6 years ago

Original report by Francisco Vargas Piedra (Bitbucket: fvargaspiedra, GitHub: fvargaspiedra).

Hi Matt,

I'm using your library for one of my text retrieval projects. I wonder why you chose the formula:

idf = log(dc / (n + 1)) + 1

Where log is actually natural logarithm. Where did you get that formula and why did you choose it?

THANKS!

stevennic commented 5 years ago

This article should answer your question.