Email Spam Blocker 2.36

Automatically detects and moves email spam from the inbox to the email spam folder. Only shows email notifications for non-spam emails.

(Please note: The app does not auto remove email spam from the inbox of Android email apps unless the inbox is set to resync with the email server. Also, be sure to choose an email spam folder on the account setup page! The auto detector requires user training, pressing thumbs up and down icons on the email spam training page.)

Filtering:

• Auto: Uses a machine learning algorithm to make intelligent email spam filtering decisions.
• Blacklisting: Define rules for emails that should be blocked. [Ex: Subject is "Viagra"]
• Whitelisting: Add email addresses to the contact app to avoid family, friends & colleagues being filtered.

Training:

• Baked Spam: The app is able to detect common spam phrases upon first use. [Ex: "Act now", "While supplies last"]
• User Training: On the history page, press the thumbs up and thumbs down icons to accept the auto spam filter's decisions.
• Adaptive: The auto spam filter will use the emails that you certify to make better decisions in the future.

Directions: 1, 2, and 3

• Setup account: Press "Accounts" to set up an email account and press 'Check connection' to confirm the email setup.
• Set spam folder: Choose an email spam folder for your account.
• Help Me Learn: Go to the training screen to confirm or reverse the decisions made by the app.

Good to Know:

• Mail Support: IMAP, IMAP IDLE, and passwordless Gmail authentication (OAuth2)
• Completely private! Your data/email never leaves your device!
• No subscription fees since the email processing is performed locally.
• POP3 is an older email protocol that does not support email folders. The app cannot move emails without email folders. However, the app can still be used with POP3 as an email notification app that only shows email notifications for non-spam emails.

The Science:

• Bayes Filter: Recursive Bayesian estimation, also known as a Bayes filter, is a general probabilistic approach for estimating an unknown probability density function recursively over time using incoming measurements and a mathematical process model.
• History: The approach to spam filtering used in this app is based on a paper called "A Plan for Spam" by Paul Graham: A Plan for Spam

Contact [email protected] for questions, suggestions, and help!