SpamCombat takes the decision about spam at the result of passing incoming emails through its filters. DNSBL filter and the Bayesian filter are used to identify spam. DNSBL filter consists in comparing the senders' IP addresses against lists of known spam databases using Public Blacklists (also called DNSBL lists). The Bayesian filter is the most powerful spam filter based on the analysis of the message content and mathematical calculation of spam. The advantage of the Bayesian filter is that the filter can be trained by each individual user by categorizing each received email as either spam or good; after the user has categorized a few emails the filter will begin to make this categorization by itself. If the filter makes a mistake, the user re-categorizes the email; the filter learns from its mistakes. The accuracy of the Bayesian filter increases with time. "Well trained" filter can determine up to 99.5% of spam emails coming into your inbox.