A malicious communication log detection method executed by a malicious communication log detection device, the method comprising:
- compressing, using a predetermined algorithm, first character strings representing characteristics of a communication log that is known to be malicious or benign, and second character strings obtained by coupling the first character strings with a character string representing a characteristic of a communication log as a classification target;
- calculating a score for determining whether the communication log as a classification target is malicious or benign based on a data size of the first character strings compressed at the step compressing and a data size of the second character strings compressed at the compressing;
- and
- determining whether the communication log as a classification target is malicious or benign based on the score calculated at the calculating and a predetermined parameter.