Site Loader

Step 2: Admin login The system will start reading themessages and attached document such as images by user’s string. It happenedwhen users send messages and input suspicious word in images the system adminwill automatically start monitoring the messages and images that containsuspicious word in short form or coded word in stored database. This section isused to monitor user’s activities and check for suspicious word modification byadding a new code word. Admin has username and password that can login andcheck all sharing details of users and information when the system flags it tobe suspicious word. Also scanning all messages and post images to identifyingsuspicious word short form or code word available, perform task of encoding anddecoding plain text hidden in images with help of cryptography algorithm.

Encrypt and decrypt the image/text.The proposed methods using cryptographyalgorithm triple DES, (3 data encryption standard) this algorithm is forencryption or decrypting the information that are sent to users. We detect thesuspicious messages / images sent from users who are already registered onsocial media. Although new users sign up themselves on the site to send shortmessages and attached documents of those who already registered and view themessages and sending request from registered users, but if they do notregisters they cannot be able to login{Rani2015}.

Triple DES algorithm used by admin toencrypt the messages or file attached, such images are sent to users or sentsome block suspicious words in short form or code words not to be access byusers. There are two procedures we used for encoding and decoding messagesshared among users are {Rani2015}:·        The process of encodingthe plaintext into cipher text / image is called Encryption.·        The process of decodingciphers text to plaintext / image is called DecryptionShort form and code word detected We used suspicious words dictionary todetect the suspicious words of all kind in any form which are not actually usedin the normal messages of communication. Admin part has a different database thatkeep up with short form of suspicious words. For example, users detail informationare: names, email, location names but for short form of suspicious word, forexample, kl (kill), att (attack), bom (bomb) and cash ($). Using WorldNetontology techniques for extracting information from unstructured data and datamining approach {Ali2013}.Code word Contrasting a few messages impactedinside same group can be recognized as code words.

In the event that same wordwas utilized by various individuals in discussion inside a gathering alongsideknown suspicious words in database, then these words are considered as codewords and furthermore added to suspicious rundown to recognize suspicious wordsin future {Thivya2015}.New suspicious word updated New suspicious words that are not as ofnow in database are established with the assistance of code words discovery techniqueand will be included back in ontology. In this manner attitude utilized here iscompletely refreshed without even a second’s pause.

This ontology refresh helpsin finding suspicious words in dynamic way and it releases time in recognizingsuspicious words in future {Thivya2015}.Pre-processing The filtering of messages and files is pre-processingin text mining approaches started by checking suspicious word in the dataset byremoving unnecessary word, check errors spelling if messages are correct. Thisstage includes text corpus consists large set of structured text messages insocial media.

Text corpus consists stop word, stemming and remove word incomputing by Natural Language Processing Algorithms.Machine Learning, NLP: TextClassificationText Classification assigns at least oneor more classes to a document as specified by their contents. Classes arechosen from a formerly established taxonomy categorization (a hierarchy systemof classifications or classes). Document classification is an issue in libraryscience for checking Text corpus database and extracting data of a fewstructured information, example of this documentations might be classified bytheir subjects or as indicated by different attribute’s, (for example, composedocument, date, year, sender and recipient details, time and so on.

There areseveral approaches of text classifications, which are as follows:

Post Author: admin

x

Hi!
I'm Dora!

Would you like to get a custom essay? How about receiving a customized one?

Check it out