A cache folder identification method and device, the method comprising: in the process of running target software, acquiring a plurality of sample values x1, x2, . . . xn of the number of files in the target folder according to a preset rule, the target folder being a folder used in the process of running the target software, n≥2; according to a preset algorithm, calculating the statistical value of the plurality of sample values, the statistical value being used to represent the degree of dispersion of the sample values; determining whether the statistical value is greater than a preset first threshold; if so, identifying the target folder as a cache folder. The method has high identification efficiency, and can continuously monitor specific software without missing a cache folder.