Privacy Preserving Spam Detection Using Hadoop
Loading...
Files
Date
2018-06-19T10:28:22Z
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Spam has become the platform of choice used by cyber-criminals to spread
malicious payloads such as viruses and trojans. In this project, we consider the problem of
early detection of spam campaigns. Existing collaborative spam detection techniques can
deal with a lot of e-mail data contributed by various sources; however, they have a
common and major problem of requiring disclosure of e-mail content. These hashes
which preserve distance are one of the common solutions used for maintaining the
privacy of the content of the e-mail while allowing the messages to get classified for
detecting the spam.
However, distance-preserving hashes are not scalable, thus making large-scale
collaborative solutions difficult to implement. To solve this, in this project, we propose a
method using Big Data which uses privacy-preserving collaborative spam detection
platform built based on a standard Map Reduce facility. It uses a highly parallel encoding
technique that enables the detection of spam campaigns in competitive times.
The evaluation of our system’s performance is done using a huge elaborate spam
base and show that our technique performs very well against the creation and delivery overhead of the current spam generation tools.
Description
Keywords
SANDEEP SK, SAMBUDDHA BISWAS, SAURABH RAJNALA, 1NH14CS168, 1NH14CS113, 1NH14CS166