Learning Loom - Tales of Technology and the Human Spirit

Learning Loom - Tales of Technology and the Human Spirit

Home
Notes
Series
Archive
About

bayesian-spam-filtering

Building a Bayesian Spam Classifier from First Principles
In 1998, Microsoft Research published a groundbreaking paper that changed how we fight spam forever.
 • Shrijith Venkatramana
Exploring Enron Spam Dataset for Training Bayesian Spam Classifier
I implemented a toy Bayesian Spam Classified in part 4 of my series on the Microsoft Research paper.
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 8
The following data table summarizes the classifier performance:
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 7
Now we are to understand a table like this:
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 6
In the Bayesian method - we start with “common sense” - execute the algorithm to “update priors” to get “posterior” result.
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 5
In the case of junk email filtering - as users and now programmers - there’s already a set of commonsensical points understood indicating higher…
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 4
Today - I looked into how to build a toy vector space from scratch, learn from it using Bayesian methods, and then classify some new toy messages as…
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 3
The paper introduces the Bayes’s theorem in brief as follows:
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 2
Bayesian Network Intuition
 • Shrijith Venkatramana
A Bayesian Approach to Filtering Junk E-Mail - Part 1
Bayesian spam filters are generally known at a popular level in knowledgeable tech circles.
 • Shrijith Venkatramana
© 2026 Shrijith Venkatramana · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture