the stood up. According to the experience of this new approach can adapt itself to the time change the strategy & 39; spammers.
the base of the principle of relying on filtering
most of the events and the probability of an event that occurred later in the suit that event occurs. inferred from the past. This approach is used to identify spam. Most of the work occurs, the text of some of the spam mail is not legitimate, then it is reasonable to suppose that email is probably spam. technology to filter mail
to
base, using the database of words that you need to generate and collect the spam from legitimate mail. Each word is assigned to calculate the probability that the probability is considered gapeunhamyeo based on how often that word occurs opposed to legitimate mail is spam. Legitimate
after and spam database created during the initial training period, the probability of that word filters to base calculations and ready for use. The new mail arrives, the most important words and the words are divided into the starter. These words from the base, the new filter to calculate the probability of whether the message is spam. If the threshold is greater than the probability of spam, 0.9, for example, was classified as spam messages.
tip! Hot g - spamcombat key to the lock using the common task allocation. For example, to assign a message marked as spam message to the suit. Nureupni to display all the features and cleaning F9. The following training simply beonebeyiseu used filters to suit your keyboard 2006, and two press F9. Note that the analysis
it is important to perform a specific distribution of spam and legitimate mail users (institutions, companies, etc.) receive, and therefore this particular person is adjusting the filter base, the company, or organization. For example, many financial institutions to receive e-mails that suit " mortgages & quot; words and will get a lot of bad judgement - if you are using an outdated spam filter to prevent. The word filtering the database to analyze the entire message is contained in the housing mortgage " & quot; conclusion, and email is spam or legitimate basis of whether a single keyword, as well as home mortgage " & quot;. . Spam is a very effective approach to filter out the base - can get spam detection rate of 99.7% more than the very low number of plausible! What benefits summary of the arrival
let base
using spam filters to catch:
1) to review all aspects of a much more intelligent approach, because the message is, as opposed to spam keywords to check on the basis of a single category word.
2) his own - to adapt - constantly learning new spam and new effective inbound mail, and spam filtering technology to adapt to the new database development.
3) sensitive to the user - and learn to understand the habits of the company& 39;s e-mail, for example, e-mail is " mortgages & quot; words are always the spam messages.
4) multiple - and the international language - any language is adapted to be used for suit. Considering the specific language-based filters are a variety of different areas, or derail the specific purpose of the words in the same language, even voice.
5) hard to fool, as opposed to keyword filters - an advanced spam filter, who wants to trick the base represents one of the few words usually used by spam, or more generally indicates that the word effective distribution (such as : A valid contact name, etc.). The latter is impossible because the spam is a profile of each recipient& 39;s e-mail to know - and this kind of spam that can never hope to collect information from all recipients.
Bookmark it:
