Sms spaming detection using NLP techniques
ABSTRACT
• In today’s digital world, Mobile SMS (short message service)
communication has almost become a part of every human life. Meanwhile
each mobile user suffers from the harass of Spam SMS. These Spam SMS
constitute veritable nuisance to mobile subscribers. Though hackers or
spammers try to intrude in mobile computing devices, SMS support for
mobile devices become more vulnerable as attacker tries to intrude into the
system by sending unsolicited messages. An attacker can gain remote
access over mobile devices. We propose a novel approach that can analyze
message content and find features using the TF-IDF(term frequency-
inverse document frequency) techniques to efficiently detect Spam
Messages and Ham messages using different Machine Learning
Classifiers. The Classifiers going to use in proposed work can be measured
with the help of metrics such as Accuracy, Precision and Recall. In our
proposed approach accuracy rate will be increased by using the Voting
Classifier.
Existing system
• The problem of sms spam detection and thread identification. The art
clustering-based algorithm are used in this work. It has two stages, in
first stages the binary classification technique such as NB, SVM, LDA
and NMF is used to categorize the sms into spam or ham sms, the
second stages sms clusters are created for ham sms using non negative
matrix factorization and K-means clustering techniques. The sms spam
detection and thread identification are used in many of sms activities
such are SMS folder classification, SMS classification and SMS thread
summarization. sms threads use two levels, the first is classification
and second is clustering. Sms threads consists of sms messages, so it
can recognize the previous communication in a message. NMF
clustering technique performs better than K-means clustering
techniques in terms of number of SMS messages participating in
threads identified.
Disadvantage
• Filtering spam messages since sms classification are
becoming more challenging due to the complexities of
the spammers. The methods of term frequency-inverse
document frequency (TF-IDF) and Random Forest
Algorithm will be applied on data and found the
accuracy among them. Only accuracy cannot
determine the performance of the algorithm. Hence
determining the precision, recall and fmeasure of the
algorithms are been observed. Performance of the
algorithm various based on the features used in the
data set.
Proposed system
• In this examination AI instrument is utilized for the analysis and
classification of the dataset. At the principal level information is assembled
from various sources to make a decent dataset of ham and spam in text
format and give that information as the input for the model. At the second
degree of the investigation we changed over the informational collection
which is prior in the text format to CSV (Comma Separated Value). At that
point pre-processing is accomplished for a superior quality info either by
removing of unrequired words or by performing stemming on them. Then
the pre-processed data information is changed into a machine readable
form or non-contextual form by changing over to vector or by doing
discretization. The labeled data is opened and the attributes are recorded.
The attributes that are utilized for the investigation intention are text and
class in this dataset. From that point forward, a classifier is applied to the
dataset we have used. Hence the information is trained utilizing the dataset.
Advantage
• we have gathered dataset then we can apply sequential steps
on the dataset first doing the exploratory data Analysis (EDA)
on the dataset then go for Test Preprocessing to clean the
message text like to remove the special symbol, convert the
text into lower case and so on. Next step to convert cleaned
text into the numerical value before to apply classifiers for
that we use TFIDF Technique to extract the feature. After the
feature extraction we apply different individual and ensemble
classifiers such as Random Forest, Bernoulli Naïve Bayes,
Support Vector Machine, Bagging Classifier, and Extra
Random Tree and then apply voting classifier to vote which is
the best individual classifier for the spam detection.
System architecture
SYSTEM SPECIFICATION:
• HARDWARE REQUIREMENTS:
• System : Pentium IV 2.4 GHz.
• Hard Disk : 40 GB.
• Floppy Drive : 1.44 Mb.
• Monitor : 14’ Colour Monitor.
• Mouse : Optical Mouse.
• Ram : 512 Mb.
SOFTWARE REQUIREMENTS:
• Operating system : Windows 7 Ultimate.
• Coding Language : Python.
• Front-End : Python.
• Designing : Html,css,javascript.
• Data Base : MySQL.
