Path: news.ccsf.jp!tomockey.ddo.jp!border1.nntp.dca.giganews.com!nntp.giganews.com!local01.nntp.dca.giganews.com!news.giganews.com.POSTED!not-for-mail NNTP-Posting-Date: Mon, 19 Nov 2007 12:34:26 -0600 Newsgroups: fj.sci.math Subject: Announcing DTREG 7 with Gene Expression Programming and Symbolic Regression From: "Phil Sherrod" <PhilSherrod@REMOVETHIScomcast.net> Date: Mon, 19 Nov 2007 18:34:25 GMT X-Newsreader: News Rover 13.1.0 (http://www.NewsRover.com) Message-ID: <j7GdnfmxPISvStzanZ2dnUVZ_rOdnZ2d@giganews.com> Lines: 74 X-Usenet-Provider: http://www.giganews.com X-Trace: sv3-npuN/fptn2aSOL1tPcV5zF5ekQ5mGWPWK1cYY676GEQzDn2EP2LfSbs/ykfLEgECIdmVCEVe8isjwjl!6hZwT73ENw124DwGrSJTaL5EdWdsufEO6rmkgolwLlebcYWLJfTOb3ECirCK2U1EtwnaK05yE7w= X-Complaints-To: abuse@giganews.com X-DMCA-Notifications: http://www.giganews.com/info/dmca.html X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.3.36 Xref: news.ccsf.jp fj.sci.math:2179 DTREG version 7 is now available with Gene Expression Programming and Symbolic Regression. DTREG home page -- http://www.dtreg.com Gene Expression Programming is a procedure that mimics biological evolution to create a computer program to model some phenomenon. The type of gene expression programming implemented in DTREG is Symbolic Regression � so named because it creates a symbolic mathematical or logical function. Gene Expression Programming is an elegant and efficient approach to genetic programming and symbolic regression. GEP was developed in 1999 by C�ndida Ferreira. Ferreira devised a system for encoding expressions that allows fast application of a wide variety of mutation and cross-breeding techniques while guaranteeing that the resulting expression will always be syntactically valid. Experiments have shown that GEP is 100 to 60,000 times faster than older genetic algorithms. DTREG provides a full implementation of the Gene Expression Programming algorithm developed by C�ndida Ferreira. Here are some of the features of DTREG�s implementation: * Continuous and categorical target variables * Automatic handling of categorical predictor variables * A large library of functions that you can select for inclusion in the model * Mathematical and logical (AND, OR, NOT, etc.) function generation * Choice of many fitness functions * Both static linking functions and evolving homeotic genes * Fixed and random constants * Nonlinear regression to optimize constants * Parsimony pressure to optimize the size of functions * Automatic algebraic simplification of the combined function * Several forms of validation including cross-validation and hold-out * Computation of the relative importance of predictor variables * Automatic generation of C or C++ source code for the functions * Multi-CPU execution for multiple target categories and cross-validation In ordinary mathematical regression, the procedure is given the form of the function to be fitted to the data. This could be a linear function for linear regression or a general mathematical function for nonlinear regression. The regression procedure computes the optimal values of parameters for the function to make the function fit a data set as well as possible, but the regression procedure does not alter the form of the function. With Symbolic Regression, the form of the function is not known in advance, and it is the goal of the procedure to evolve a function that will fit the data. In addition to evolving mathematical functions, symbolic regression can be used to develop logic circuits using AND, OR, NOT, XOR and other logical functions. In addition to Gene Expression Programming, DTREG provides these other types of models: * Support vector machine * Multilayer perceptron * RBF neural networks * Decision trees * Boosted decision trees * Decision tree forests * Probabilistic and General regression neural networks (PNN/GRNN) * Cascade correlation networks * Linear discriminant analysis * Logistic regression You can download a demo copy of DTREG from http://www.dtreg.com/DownloadDemo.htm Detailed information about Gene Expression Programming can be found at http://www.dtreg.com/gep.htm -- Phil Sherrod (PhilSherrod 'at' comcast.net) http://www.dtreg.com (Decision trees, Neural networks and SVM modeling) http://www.nlreg.com (Nonlinear Regression)