Email White Papers

An Evaluation of Naive Bayesian Anti-Spam Filtering

Overview It has recently been argued that a Naive Bayesian classifier can be used to filter unsolicited bulk e-mail ("Spam"). This paper conducts a thorough evaluation of this proposal on a corpus that they make publicly available, contributing towards standard benchmarks. At the same time they investigate the effect of attribute-set size, training-corpus size, lemmatization, and stop-lists on the filter's performance, issues that had not been previously explored. After introducing appropriate cost-sensitive evaluation measures, they reach the conclusion that additional safety nets are needed for the Naive Bayesian anti-spam filter to be viable in practice.

Further White Paper Details
PublisherDemokritos File FormatPDF
Date PublishedMay 2000
FormatWhite Papers   
Topics

Email as Evidence

Discover the 12 steps to ensuring good evidential quality of data. Learn why today's litigious and regulatory environment means your email archive must support discovery and disclosure requests. Written in...

Video case study

Duncan Scott, CIO of DTZ, talks about their global email project and how Mimecast has helped them with their challenges operating as a multinational business

Whitepaper: Dell Proconsult Deduplication Services

Expert guidance and a proven process for reducing storage requirements and improving.

The Essential Guide: Real-Time High Availability for Exchange - Replicate Exchange Data for Improved Resiliency

Business email has become so important that most organizations consider achieving the highest possible availability a must. How can you measure—and improve—your email systems' availability and resiliency?

Streamline core business processes with IBM WebSphere MQ

In today's constantly changing business environment, it is essential for applications and business processes to remain connected and integrated. View this interactive PDF and see how IBM WebSphere® MQ provides...


Quick Sitemap Links: