Content Base Mail Classification | Computer Science Project Topics

Published on Nov 30, 2023

Abstract

Any inbox can be classified based on the contacts into four classifications, namely Private, Public, Newsletter and Anonymous. Every user has the right to choose their keyword (a semi-private password). Those contacts who know the user’s keyword will be classified as private contacts and those who do not are classified as anonymous contacts. A contact can be classified as public or private, upon verification of an anonymous contact. Any newsletter or group mails are classified into newsletter contacts. It is highly likely that the rest are junk mail or spam.

In order to make sure that spam mails are at bay, a simple yet highly effective and efficient mail classification system has been developed upon constant testing. The system involves classification of mails based on user’s contacts. This way any mail from a contact whom the user knows very well is are displayed.

Spamming

Spamming is currently an active area of research today. From elementary word based spam filters to complex and efficient filters like Bayesian filters. The design of these filters has always been the center of argument for years. This project involves developing of an Electronic alert based mail client developed based on A novel two-tier spam filter. This novel approach defines spam prevention as a classic puzzle in the research area of network security. The conventional spam filtering techniques still result to high false positives and have weak on-line processing ability. This paper presents a novel Two-Tier Spam Filter (TTSF) on the basis of analysis over the conventional anti-spam filtering techniques. TTSF on-line filters spam by using URL’s and off-line filters spam using digest-based approach.

Experimental evaluations from the constructed prototype based on TTSF demonstrate it can significantly raise the filtering accuracy, effectively reducing false positives and can be applied to online processing and heavy traffic environment by reducing the computational cost that the state-of-the-art techniques. In Contact based electronic mail classification, an elementary filter based on the keyword from the user is used to classify mails that arrive in the INBOX. The classification categories [folders] are Private, Public, Newsletter and Anonymous. The user account has a unique Mail ID and a keyword is chosen for the corresponding ID. A keyword is a Semi-private password. It can be modified later like password.

Proposed System

1) The keyword is given along with the Mail ID, only to those who are known well to the user [eg. Relatives / Friends / etc..].

2) When a mail with the keyword arrives from such very well known persons, it is marked as PRIVATE.

3) When a mail from people who doesn’t know the keyword, is automatically sent a CAPTCHA code / Question for verification.

4) Upon verifying the CAPTCHA / answer for the Question, the person could either fall into PUBLIC / PRIVATE.

5) If the sender is well known he is marked PRIVATE. If the sender is not well known he is marked PUBLIC.

6) If the user registers to receive Newsletters, the sender’s ID which is provided during registration, is placed by the user in NEWSLETTER category.

Advantages of Proposed System

1) With the implementation of Keyword association for an user account, the user will be provided with his private space for private mails.

2) Upon implementation of personalized keyword, the system is 100% efficient against spam / junk mails.

3) No false positives are registered. The user is displayed mails only from his reliable contacts in his private mailbox.

4) Mails from those anonymous contacts who have been verified, yet not reliable will be placed in the public inbox. This offers a second degree of protection from spammers.

5) Newsletters / Groups who are reliable and offer bulk mailing to user’s will be available in the Newsletter inbox. So, mails from such reliable sources are displayed without verification.

Content Base Mail Classification

Applications

1) Contact based mail classification can be implemented on any type of existing mail service. Independent implementation across public or private mail service providers is feasible.

2) This client can also be used to implement on a particular mail account or an user using free POP and SMTP services offered by mail service providers.

3) A personalized mail client can be developed to incorporate all the existing features from popular mail clients and hence offer a completely free to use.

4) There is no need to modify the protocols in-order to implement this system, as sending and retrieving mails are the only abstracted core components of this mail client.

Highlights of the Project

All though several mail clients exits today, it is highly likely that a false positive rises. A simple mail from a new unknown user with a subject “hello” and mail content reading “hi”, will be placed under spam folder by many mail service providers for the reason that their spam filtering technique classifies a mail to a spam when the user is new and when the number of words in the subject or content is very less. The best and most efficient way to classify a mail in the user’s private space would be to give users their own private space in his public mail space.

1) The system can be used to integrate features in a single plug-in and can be used as an add-on for famous or feature rich mail clients that already exist [Mozilla Thunderbird etc..]

2) The system can be either integrated into an already existing mail service provider.

3) Also, the system can be used by an individual user on a web hosting service for much secure mail space, only when the user has POP and SMTP service.