摘要 |
A method and a system for detecting spam mails with a character input pattern are provided to determine/delete spam documents without checking each document by extracting the character input pattern from the document registered by a user, calculating a frequency of the document including each character input pattern, and determining whether the same pattern is repeated in each document based on the frequency. A character input pattern extractor(610) extracts the character input pattern from the document registered by the user according to a predetermined rule. A frequency calculator(620) calculates the frequency of documents including each extracted character input pattern. A pattern database(630) stores the frequency of each character input pattern. A spam pattern extractor(650) extracts a spam pattern among the character input patterns based on the frequency. A document processor(640) automatically processes the document including the extracted spam pattern. The pattern database stores a document ID in association with the character input pattern. A document ID identifier(660) identifies the document ID associated with the spam pattern by referring to the pattern database. The document processor reports the document corresponding to the document ID to a spam detection system manager. |