Vowpal Wabbit (Fast Learning)
This is a project started at Yahoo! Research and continuing at Microsoft Research to design a fast, scalable, useful learning algorithm.
VW is the essence of speed in machine learning, able to learn from terafeature datasets with ease. Via parallel learning, it can exceed the throughput of any single machine network interface when doing linear learning, a first amongst learning algorithms.
We primarily use the wiki off github. A few useful starting points are:
Open-source toolkit for fast learning (Agarwal et al., 2011)
Research project initiated by John Langford (Yahoo, Microsoft)
Implemented in C++, (install: sudo apt-get install vowpal-wabbit)
Supports multiple supervised/semi-supervised methods:
Classification (both binary and multi-class), regression, active learning for both regression and classification
Matrix factorization (sparse matrix SVD) Neural net (with user specified layer count) Searn (Search and Learn)
Latent Dirichlet Allocation (LDA)
from Dr. Jochen L. Leidner-Data Science: A Compact Introduction