Fingerprint-based data classification

A computer program product comprising one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to cause a computing device to:

• generate, based on predetermined rules, a fingerprint of a data column in a data set to be classified, the fingerprint comprising column dimensions, wherein each of the column dimensions is assigned an attribute representing a characteristic of data in the data column;
• determine whether the fingerprint matches one or more target fingerprints by comparing the fingerprint to the target fingerprints, wherein each target fingerprint is associated with a class and includes class dimensions, each class dimension is assigned an attribute representing a characteristic of data in the class, and each of the one or more target fingerprints further include a weight assigned to each of the class dimensions based on a frequency distribution of each of the class dimensions;
• in response to determining that the fingerprint matches the one or more target fingerprints, assign one or more classes to the data column based on the one or more target fingerprints, thereby generating classified data;
• and
• in response to determining that the fingerprint does not match the one or more target fingerprints, perform a full data classification of the data column by comparing data of the data column to multiple data classes to determine whether at least one of the multiple data classes is associated with the data column.

A system comprising:

• a processor, a computer readable memory, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to cause a computing device to:
• generate, based on predetermined rules, a fingerprint of a data column in a data set to be classified, the fingerprint comprising column dimensions, wherein each of the column dimensions is assigned an attribute representing a characteristic of data in the data column;
• determine whether the fingerprint matches one or more target fingerprints by comparing the fingerprint to the target fingerprints, wherein each of the target fingerprints is associated with a class and includes class dimensions, each class dimension is assigned an attribute representing a characteristic of data in the class, and each of the one or more target fingerprints further include a weight assigned to each of the class dimensions based on a frequency distribution of each of the class dimensions;
• in response to determining that the fingerprint matches the one or more target fingerprints, assign one or more classes to the data column based on the one or more target fingerprints, thereby generating classified data;
• and
• in response to determining that the fingerprint does not match the one or more target fingerprints, perform a full data classification of the data column by comparing data of the data column to multiple data classes to determine whether at least one of the multiple data classes is associated with the data column.