Page 55 - ASBIRES-2017_Preceedings

AVOID EMAIL SPAMMING BY SERVER AUTHENTICATION AND SAILJS ORM
2.2 Naïve Bayes Classifier Method
In 1998, the Naïve Bayes classifier was
proposed for the recognition of spam. The
Bayesian classifier is working on the
dependent events and the probability of an
event occurring in the future that can be
detected from the previous occurrence of the
same event.
This technique can be used to classify
spam emails; Probabilities of words play
the main role here. If the total word
probabilities exceed a certain limit, the filter
will mark the email in any of the categories.
Here, only two categories are needed: spam
or ham. Almost all statistical-based spam
filters use the Bayesian probability
calculation to combine individual token
statistics with a global score. he following
equation is used to calculate spam
probability.
Where, S spam (T) and C Ham (T) are the
number of spam or ham messages that
contain the T token, respectively. To
calculate the possibility of an M message
with tokens {T1... TN}, it is necessary to
combine the spamming of the individual
token to evaluate the general message of
spamminess.
A simple way to make classifications is
to calculate the spamminess product of an
individual token and compare it with the
individual token product (Rao & Reiley,
2012).
2.3 Artificial Immune System Classifier
Method
Biological immune system has
succeeded in protecting the human body
against a wide variety of foreign pathogens.
One role of the immune system is to protect
our body from infectious agents such as
viruses, bacteria, etc. On the surface of these
agents are antigens that allow the
identification of the invading agents,
45
therefore provoking an immune response
Recognition in the immune system is
performed by lymphocytes.
3 METHODOLOGY
Figure 1: System diagram
3.1 Preprocessing Email
The content of the email is received
through our software. The information is
then extracted as mentioned above. Then the
extracted information (feature) is stored in a
corresponding database. Each message
became a function Vector with 21700
attributes (this is approximately the number
of different words in all corpus messages).
An attribute n was set to 1 if the
corresponding word was present in a
message and in 0 otherwise.
This feature extraction scheme was
used for all algorithms used for spam
filtering.
3.2 Description of the Extracted Feature
Email Sending feature mainly consisted
of Lossless data compression algorithms
usually exploit statistical redundancy to
represent data without losing any
information, so the process is reversible.
Lossless compression is possible because
most real-world data show statistical
redundancy.
For example, an image may have color
areas that do not change over several pixels;
Instead of encoding "red pixels, red pixels",
the data can be encoded as "279 red pixels".