| 主权项 |
1. A method for protecting a mail server, comprising:
detecting, via a processor, the mail server is reaching a processing limit pertaining to a load level experienced by the mail server; and selectively limiting connections to the mail server, via the processor, from a plurality of source nodes based on a spam index associated with each of the plurality of source nodes, wherein the spam index is associated with an internet protocol address of each of the plurality of source nodes, wherein the spam index is converted into a priority index, wherein the priority index indicates a probability of a connection to a respective internet protocol address being dropped, wherein the spam index comprises a ratio of spam mail to non-spam mail, wherein the spam index is generated by a classifier, wherein the classifier is trained using a plurality of feature vectors, wherein each of the plurality of feature vectors is derived from flow level statistics, wherein each of the plurality of feature vectors is further derived from content level statistics of a corresponding source node of the plurality of source nodes, wherein the classifier is further trained using a plurality of spam indices associated with each of the plurality of feature vectors, wherein the plurality of spam indices associated with each of the plurality of feature vectors is generated from spam filter logs provided by a spam filter. |
