Welcome to CrfCom's Home!

      Links:   CrfCom Project Home   |     Download CrfComLib   |     Browse Documentation

CRF and CrfCom


    CRF ( Conditional Random Field ) is a probabilistic model used in labeling or parsing sequential data. It has been proved to be effective in fulfilling such tasks as Named Entity Recognition, Information Extraction and Text Chunking.

    CRF++(CrfPP) is a well-versed C++ implementation of CRF by Taku Kudo. It provides a simple, yet powerful  and portable tool for researchers and developers who wish to try and test their research scheme based on CRF.

    There are  other open-source CRF implementations in C++ available (eg. hCRF, CRFsuite etc.), but most of them are provided as standalone tools and are not intended to be readily used as a component for GUI applications.

    CrfCom is a COM library based on CrfPP. It is a not only a wrapper of CrfPP, but  also provides a number of extra funtional features that CrfPP and other similar tools have not  been equipped with.


Main Features



How to Use


    CrfCom is a very easy-to-use component for clients implemented in different languages like C#,VB,C++ and other script languages.

    To start using CrfCom, please refer to the Getting Started section of the online documentation.   



  1. J. Lafferty, A. McCallum, and F. Pereira. Conditional random fields: Probabilistic models for segmenting and labeling sequence data, In Proc. of ICML, pp.282-289, 2001
  2.  Taku Kudo, CRF++: Yet Another CRF toolkit V0.51,  Jul 11 2008




© HBCHEN. All Rights Reserved.