Spam Not any more Detecting spam emails using

  • Slides: 8
Download presentation
Spam? Not any more !! Detecting spam emails using neural networks ECE/CS/ME 539 Project

Spam? Not any more !! Detecting spam emails using neural networks ECE/CS/ME 539 Project presentation Submitted by Sivanadyan, Thiagarajan

Importance of the topic u u Spam is unsolicited and unwanted emails Wastage of

Importance of the topic u u Spam is unsolicited and unwanted emails Wastage of bandwidth, storage space and most of all, recipient’s time Goals of the Anti-spam Network u u u Reliably block spam mails Should not block any non-spam mails, but can allow few spam mails to slip through Adapt to the specific types of messages

Input Features – Data Set Original data set: 57 input attributes u Output attribute:

Input Features – Data Set Original data set: 57 input attributes u Output attribute: 1 (for spam) 0 (for nonspam) u Inputs derived from email content u Attributes indicate the frequency of specific words and characters u Examples: ‘credit’, ‘free’ (in spam) ‘meeting’, ’project’, (in nonspam) u

Preprocess the data Choose only the inputs which differ for spam and non-spam mails

Preprocess the data Choose only the inputs which differ for spam and non-spam mails u Two reduced data sets are obtained (21 Inputs and 9 Inputs) u The data is made zero mean, unit variance (4025 Input Vectors) u Split the data into two independent training and testing data sets u

MLP Implementation Learning by back propagation algorithm u Using complete data set u •

MLP Implementation Learning by back propagation algorithm u Using complete data set u • Poor performance (Classification rate: 63. 2%) • Classified most of the mails as non-spam u Using reduced data set (Inputs – 21) • • • Good performance (Classification rate: 93. 8%) All the non-spam is detected Optimal MLP Configuration: 20 -10 -10 -10 -7

Cross Validation u Using reduced data set (Inputs – 9) • Good performance (Classification

Cross Validation u Using reduced data set (Inputs – 9) • Good performance (Classification rate: 92. 1%) • Nearly all the non-spam is detected • Optimal MLP Configuration: 20 -10 -10 -8 u Using Cross - Validation • • Negligible improvement in performance Since all the data is derived from the same source, cross validation offers no advantage

Inference of the results u u Larger number of inputs does not necessarily improve

Inference of the results u u Larger number of inputs does not necessarily improve the performance It is important to remove redundant and irrelevant features There is no optimum MLP configuration for all inputs – need to adapt depending on the email content A combination of other types of spam filters along with neural networks can be used

Conclusion u u Neural networks are a viable option in spam filtering A number

Conclusion u u Neural networks are a viable option in spam filtering A number of heuristic methods are being increasingly applied in this field Need to exploit the differences between spam and ‘good’ emails Further opportunities • Data sets from different sources need to be used for training • Fuzzy logic and combinational algorithms can be used in this application

EMAIL DAN ANCAMANNYA SPAM SPAM FILTER SPAM SpamEMAIL DAN ANCAMANNYA SPAM SPAM FILTER SPAM Spam
Spam Detection Kingsley Okeke Nimrat Virk Spam emailsSpam Detection Kingsley Okeke Nimrat Virk Spam emails
More more I want more More More MoreMore more I want more More More More
Spam Overview What is Spam Spam is unsolicitedSpam Overview What is Spam Spam is unsolicited
Introduction Email Spam IM Spam Text Spam BlogIntroduction Email Spam IM Spam Text Spam Blog
Spam Co to jest SPAM SPAM to elektroniczneSpam Co to jest SPAM SPAM to elektroniczne
Spam and EMail Spam Spam is unwanted emailSpam and EMail Spam Spam is unwanted email