IMPLEMENTING FILTERED WALL IN ONLINE SOCIAL NETWORKING SITE SUPERVISOR Mr.DHANASEKARAN S Assistant professor Department of Computer Sc ience and Engineering DONE BY MANASY M(211611104076) NIVEDHITHA R(211611104092) RANJINI PRIYA R(21161110411 0)
IMPLEMENTING FILTERED WALL IN ONLINE SOCIAL NETWORKING SITE
SUPERVISOR Mr.DHANASEKARAN S Assistant professor Department of Computer Science and
Engineering
DONE BY MANASY M(211611104076) NIVEDHITHA R(211611104092) RANJINI PRIYA R(211611104110)
ABSTRACT
The major issue in today's On-line Social Networking is to give users the a
bility to control the messages posted on their own private wall and to avoi
d that unwanted messages being posted. Online social networking provide l
ittle support to this requirement. We propose a system allowing online soci
al networking users to have a direct control on the messages posted on thei
r walls. This is achieved through a flexible system, that allows users to cust
omize the filtering criteria to be applied to their walls.The present work is t
o experimentally evaluate an automated system called Filtered Wall,able to
filter unwanted messages from user wall.
PROBLEM STATEMENT
As more and more people are spending increasing amounts of time
on
social networking sites there is a growing concern for the privacy and
legal rights surrounding them.
This work provides a comprehensive solution for the privacy, and
security trends associated with social media.
But, like anything, as social networking sites become more popular
the risks that stem from them increase and the need for new and
updated security becomes necessary.
These sites also state that they will not notice or compensate the user
if they choose to take actions on their submitted content. In order to
minimize the risks associated with this control of a social networking
site, users should review and should take caution in what they post.
EXISTING SYSTEM
• Today online social networking provide very little support to prevent
unwanted messages on user walls. For example, Face book allows users to
state who is allowed to insert messages in their walls (i.e., friends, friends
of friends, or defined groups of friends).
• It is not possible to prevent undesired messages, no matter of the user who
posts them.
• No content based preferences are supported and therefore it is not possible
to prevent undesired messages such as political or vulgar ones, no matter
of the user who posts them.
DISADVANTAGES OF EXISTING SYSTEM
• No content-based preferences are supported and therefore it is not posible
to prevent undesired messages, such as political or vulguar ones,no matter
of the user who posts them.
• This is because wall messages are constituted by short text for which
traditional classification methods have serious limitations since short text
do not provide sufficient word occurrences
PROPOSED SYSTEM
• The aim of the present work is to propose and experimentally evaluate an
automated system, called Filtered Wall, able to filter unwanted messages
from online social networking user walls..
• This system is to automatically filter unwanted messages from online
social networking user walls on the basis of message content
characteristics.
LITERATURE SURVEY
1.Title : Content-based Book Recommending Using Learning for Text
Categorization
Author : Raymond J.Mooney , Loriene Roy,August 1999
Recommender systems improve access to relevant products and
information by making personalized suggestions based on previous
examples of a user's likes and dislikes.
Advantages:
• This approach has the advantage of being able to recommended
previously unrated items to users with unique interests using ML.
Disadvantages:
• Users have to select productive strategies for selecting good examples .
2.Title: Machine Learning in Automated Text Categorization
Author : Fabrizio Sebastiani,October 2001
Automated categorization of texts into predefined categories is done by a
general inductive process that automatically builds a classifier by learning,
from a set of preclassified documents.
Advantages:
• The advantages of this approach over the knowledge engineering approach
are a very good effectiveness, considerable savings in terms of expert
manpower, and straightforward portability to different domains.
Disadvantages:
• Three different problems namely document representation,classifier
construction and classifier evaluation.
3.Title : Automated Learning Of Decision Rules for Text Categorization
Author: Chidanand Apte, Fred Damerau, Sholom M. Weiss,1994
This method is to automatically discover classification patterns that can be
used for general document categorization or personalized filtering of free
text.
Advantages:
• Shows a large gain performance.
Disadvantages:
• Using dictionaries of single word does not mean that the best solution
ignores phrases and combinations of words.
4.Title: Combining Provenance with Trust in social Networks for
Semantic Web Content Filtering.
Author : Jennifer Golbeck
An algorithm for inferring trust relationships using provenance
information and trust annotations in Semantic Web-based social networks.
Advantages:
• Film trust is presented as an application and the results obtained with
FilmTrust illustrate the success that can be achieved using this method.
Disadvantages:
• Networks are different.Depending on the subject about which the trust is
being expressed,the user communityand effect of these properties of trust
can vary.
5.Title : RCV1 A New Benchmark Collection for Text Categorization Research
Author : David D.Lewis, Yiming Yang, Tony G.Rose,Fan Li,2004
Reuters Corpus Volume I (RCV1) is an archive of over 800,000 manually categorized
newswire stories.This provides a benchmark data on all categories .There are 103
Topic categories, 101 with one or more positive training examples on training set.
Advantages:
• Incorporated supervised learning approaches on the RCV1 data, to provide
benchmark and a check that corrections to the data did not introduce any new
anomalies.
Disadvantages:
• The number of duplicates,foreign language documents and other anamolies is
problematic and depends on the questions the researchers use.
SYSTEM ARCHITECTURE
MODULES
• LOGIN AUTHENTICATION AND REGISTRATION:
Login:
The login module presents visitors with a form of username and password
fields.If the user enters valid username and password then they will be
granted access to additional resources on the website.
Registration:
It is the ability to create new users. New users have to give their details.
Having their account gives many features, including more editing options
and user preferences.
Login and Registration:
• PROFILE GENERATION :
User’s profile details like profile name, display picture and status are
entered by the user which gets stored in the database.
Authorized users once logged into their profile can see their details
and user can edit their profile details which gets updated.
New user Existing User
• SEND FRIEND REQUEST:
In this module user select friend to send request and can later cancel
it if they wish to.
The other user can accept or deny the friend request.
Send Request
Cancel Request
Accept or Ignore Request
•ACCEPT FRIEND REQUEST:
In this module users add new friends and view their profile details.
Logged users can see their friend list and if they wish can add friends.
They can post messages in the wall of the user who has accepted their
friend request.
•POST STATUS:
In this module user can post any post in public wall, and any friend of
user can post on the user wall.
If the posted content is postable message the content gets posted on
the user wall.
User can view their recent post and can remove it if they wish so.
View Recent posts Posted by the user
Posting in User's Friend's wallPosting in User's wall
• FILTERING TEXT :
This module manages posting comments in the user status box.
Each non postable content has an alert meassage denying the posting
of message in user's wall.
Posting Content Filtering the Post
CONCLUSION
In this work, we have presented a system to filter undesired messages
from OSN walls. The system exploits a soft classifier to enforce
customizable content-dependent filtering method. This work is the first
step of a wider project .The early encouraging results we have obtained on
the classification procedure prompt us to continue with other work that will
aim to improve the quality of classification.
FUTURE WORK
In particular, future plans contemplate a deeper investigation on two
interdependent tasks. The first concerns the extraction and/or selection of
contextual features that have been shown to have a high discriminative
power. The second task involves the learning phase. Since the underlying
domain is dynamically changing, the collection of pre-classified data may
not be representative in the longer term.. Additionally, we plan to enhance
our system with a more sophisticated approach to decide when a user
should be inserted into a BL.
THANK YOU