The "Spam System" in MailDroid is a subscription based service that uses a unique set of algorithms to mark mail as spam in your mailbox (IMAP/POP3/Exchange). There is a trial period which allows you to try out the add-on first.

There are a number of ways to look and detect spam in mail systems. The most important thing about the algorithm used in MailDroid is that it needs to learn. You must train it by marking things as spam, once trained (which may take 2 days or longer depending on how many emails you get), it will start getting better at detection rates. Mail identified as spam will get a red spam label and you can filter for spam and manually move it to a spam folder (coming soon will be an additional rule to create to move it automatically). The learning approach is very important and will result in a very robust and efficient anti-spam approach. Having said all the above, please be aware that with all spam systems, there is a small percentage of false positives where even when trained for some time, a good email may be accidently classified as a spam mail.

Starting with version 2.90, the spam option in MailDroid will now do one of 2 things:

The full features in the spam system include:

Wildcards can also be added to black/white lists so that you can easily capture an entire domain or part of a name. Some examples are given below:

In all cases, the black list will have the priority. Meaning that if you blacklisted *@gmail.com, but whitelisted joe@gmail.com, the email will be considered spam. Also, if you blacklisted joe* and whitelisted joe@gmail.com, your email will also be considered spam. The priority of defining a spam email will always fall on the blacklist.

To summarize, the techniques currently used by many anti-spam systems are static lists or english keywords, meaning that spammers who examine the latest anti-spam filtering techniques simply find ways to avoid them. The approach taken by MailDroid is dynamic and will learn and continue to change over time based on your habits.

In the future, more options may be added to this feature.