"In the past few years, we've seen an alarming increase in the number of cybercrimes involving anonymous e-mails … (that) can transmit threats or child pornography, facilitate communications between criminals or carry viruses," said Benjamin Fung, a data mining expert and professor of Information Systems Engineering at Concordia University in Montreal, the school said Tuesday.
Fung and his team created a way to locate an anonymous sender by using speech recognition and data mining, which involves gleaning useful, previously unknown information from a large volume of raw data.
An anonymous e-mail suspect may be identified through patterns -- misspellings, grammatical errors, the use of only lower-case letters, etc. -- found in other e-mails written by the suspect, the school said. Investigators would then filter out any of these patterns also found in the e-mails of other suspects.
The remaining frequent patterns amount to the suspect's "write-print," which is unique and similar to a fingerprint.
"Using this method, we can even determine with a high degree of accuracy who wrote a given e-mail, and infer the gender, nationality and education level of the author," Fung said. "For evidence to be admissible (in a court of law), investigators need to explain how they have reached their conclusions. Our method allows them to do this."