Eureqa | Cornell Computational Synthesis Laboratory     Eureqa is a software tool for detecting equations and hidden mathematical relationships in your data. Its primary goal is to identify the simplest mathematical formulas which could describe the underlying mechanisms that produced the data.     http://ccsl.mae.cornell.edu/eureqa     tags: data visualization viz ai machinelearning ml Distilling Free-Form Natural Laws from Experimental Data (Ge...     For centuries, scientists have attempted to identify and document analytical laws that underlie physical phenomena in nature. Despite the prevalence of computing power, the process of finding natural laws and their corresponding equations has resisted automation. A key challenge to finding analytic relations automatically is defining algorithmically what makes a correlation in observed data important and insightful. We propose a principle for the identification of nontriviality. We demonstrated this approach by automatically searching motion-tracking data captured from various physical systems, ranging from simple harmonic oscillators to chaotic double-pendula. Without any prior knowledge about physics, kinematics, or geometry, the algorithm discovered Hamiltonians, Lagrangians, and other laws of geometric and momentum conservation. The discovery rate accelerated as laws found for simpler systems were used to bootstrap explanations for more complex systems, ...     http://www.sciencemag.org/cgi/data/324/5923/81/DC1/1     tags: physics machinelearning genetic pendulum Python nearest neighbors binary classifier | This Number Cru...     This time we're going to build a nearest neighbors classifier.     http://blog.smellthedata.com/2009/06/python-nearest-neighbor...     tags: python machinelearning knn kdtree scipy Netflix prize tribute: Recommendation algorithm in Python | ...     In honor of the prize barrier being broken, I put together a little implementation of an early leader's approach to the problem. They experimented with several different approaches, but the one I use the most is the original probabilistic matrix factorization (PMF) approach. To see all the details, including how it performs on the full Netflix problem, see Russ and Andrei's paper.     http://blog.smellthedata.com/2009/06/netflix-prize-tribute-r...     tags: python netflix ml machinelearning http://blog.josephwilk.net/workspace/python/lsa.py     python Latent Semantic Analysis     http://blog.josephwilk.net/workspace/python/lsa.py     tags: python lsa lda svd scipy machinelearning Netflix Update: Try This at Home     http://sifter.org/~simon/journal/20061211.html     tags: netflix machinelearning Similar software - Mallet     http://mallet.cs.umass.edu/index.php/Similar_software     tags: ai clustering learning machinelearning