Term Frequency - Inverse Document Frequency (TF-IDF)

[cesarsouza]

It would be a nice addition for those using Accord.NET in text applications, specially now that more linear optimization algorithms are available.

Kory Becker has created a nice implementation (under a compatible license) which can be used as a basis for the implementation. The current code can be found here:

https://github.com/primaryobjects/TFIDF/blob/master/TFIDFExample/TFIDF.cs

However, it seems the implementation needs a stemmer. In this case, the stemmer could be incorporated in the project by specifying a new ITextStemmer interface. Different stemmers could than be created using the Snowball project:

https://github.com/cesarsouza/snowball

It would be cool to add a new text generator in Snowball for C#. It shouldn't be that difficult given that there are working Java generators available.

[cesarsouza]

Snowball generators have been added in commit 1896fe0

[cesarsouza]

Fixed in 3.5.0.