Brilliant idea, that's cool. Go for it! I remember some ANN code in the vaults of parnassus, math. section. It was a rather basic backpropagation implementation. Lars Brad Clements wrote: > > Howdy, > > I've been saving up all the spam messages I get for the past two months. I > have about 1869 spam messages saved. > > Now I'd like to develop a neural net based filter for my email program and > train it to recognize these messages as spam. > > I don't know much about neural networks, they just seemed to be the best way > to go to solve this problem. Spam messages have particular characteristics > that I can recognize (as a human), and I'm hoping to train a program to do > the same. > > Some indicators of spam are > > All caps in subject > keywords in body, like sex, viagra, penis, money, income, earn, free, free, > free > "weird" looking from address (username has many mixed digits or few vowels) > very long messages > > (lots of other things) > > None of these items by themselves are clear indicators, so I think a neural > network might be good at recognizing a spam pattern. > > Anyone have any Python code or suggestions how to proceed? > > -- > Brad Clements, > bkc at murkworks.com > > -- > http://mail.python.org/mailman/listinfo/python-list -------------- next part -------------- A non-text attachment was scrubbed... Name: vonWedel.vcf Type: text/x-vcard Size: 342 bytes Desc: Card for Lars von Wedel URL: <http://mail.python.org/pipermail/python-list/attachments/20010417/22858def/attachment.vcf>
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4