Movatterモバイル変換
[0]ホーム
Any Neural Net code in Python? I want to filter out spam email
s713221 at student.gu.edu.aus713221 at student.gu.edu.au
Fri Apr 20 08:08:17 EDT 2001
Romuald Texier wrote:>> Ken Seehof wrote:>> > Excellent idea, Dan. That's conveniently sidesteps the most difficult> > issue: getting the neural network to actually come up with linguistic> > rules. Once an intelligent human specifies the set of rules, the neural> > net should have no difficulty coming up with an optimal non-linear> > function of pre-processed features (i.e. the "rules") to identify spam.> > Analysis of the weights after training will help remove rules that turn> > out to be irrelevant.>> Wouldn't decision tree or other rule inference algorithms be more accurate> than neural networks for that kind of machine learning ? Moreover, neural> nets are "black box" : you do not get a logical rule set (that may be> edited by humans or exchanged) but an ugly matrix of floats...>> --> Romuald TexierNot for all neural net algorithms. There are a variety of neural netalgorithms that, when backpropagation is applied to a node, it's closestneighbour also recieves a little bit of this weighting adjustment aswell. What you end up with is a grouping of similarly behaved nodes,which should be easier to extract rule sets from.However, if a simple weighting of rules are all that one is after, maybea genetic algorithm is all that is required, which does give the benefitof simple rule extraction by humans. Give it an ~ 20 sized population ofweightings, apply stociastic propagation with all the mutation,cross-over functions etc., and let the population climb to an optimalpeak. Of course, neural nets have the advantage that they include theposibility of non-linear behaviour. To include non-linearity for thegenetic algorithm, you'd need to do things like this:f_n is the n'th filtering rulew_n is the linear weighting associated with this rulew_nm is the bilinear weighting associated with rules n and mAnd run the genetic algoritm for this, with penalties for using w_nm's.Only hassle is, for a rule set of size N, a population that tries tosolve a complete solution has to solve for N + N^2 (allowing for w_nnvalues) weightings. Maybe a genetic algorithm which allows incomplete but expandablesolutions, that penalises for computational size? Everything I wrotebeyond this point was too vague to include so I'm shutting up now.____________snip___________________Joal Heagney/AncientHart
More information about the Python-listmailing list
[8]ページ先頭