Treeler is an open-source c++ library of structure prediction methods, focusing on Natural Language Processing tasks like tagging and parsing. It is released under the GNU-GPL.
Treeler implements a framework for linear structure prediction. The main features of the framework are:
Structured prediction models take the form of factored predictors. A central component behind a structured prediction model is a factorization or decomposition of structures into "parts". The model is then defined according to this factorization.
The library provides learning algorithms that are generic with respect to the particular factorization employed by a model. So far, the library implements learning algorithms for classification that have been ported to structure prediction. This includes Perceptron, log-linear models, and max-margin methods.
The library provides standard factored models for multiclass classification, sequence tagging and dependency parsing.
A typical Linux box with usual development tools: bash, make, and a C++ compiler with basic STL support.
Enough hard disk space (about 120Mb)
Some external libraries are required to compile FreeLing:
- libpcre (version 4.3 or higher): Perl C Regular Expressions. Included in most usual Linux distributions. You'll need binary and development packages.
- libdb (version 4.1.25 or higher): Berkeley DB. Included in all usual Linux distributions.
If you're still curious, you can get a copy of treeler from our svn repository:
For an overview of Treeler, check the video presentation at WAPA'2011 http://videolectures.net/wapa2011_carreras_treeler .