Automatic privacy-aware machine learning method and apparatus

A non-transitory computer-readable storage medium tangibly encoded with computer-executable instructions that when executed by a processor associated with a computing device perform a method comprising:

• generating, using an unencoded query and an encoding scheme, an encoded query for searching a plurality of records encoded using the encoding scheme, each record of the plurality of records having a plurality of fields,
• the encoding scheme comprising a plurality of encoded value sets corresponding to a plurality of unencoded value sets from the plurality of records, each encoded value set of the plurality of encoded value sets corresponding to a respective one of the plurality of fields,
• an encoded value set, of the plurality of encoded value sets, corresponding to the respective one of the plurality of fields comprising an encoded value for each unencoded value in the unencoded value set corresponding to the respective one of the plurality of fields,
• the encoding scheme being used to encode each of a number of fields, from the plurality of fields, found in the unencoded query, and
• for a respective field of the number of fields found in the unencoded query, the encoding scheme comprising replacing the respective field's unencoded value with a corresponding encoded value from the encoded value set corresponding to the respective field;
• and
• generating a search result based on the plurality of records using the encoded query to search the plurality of encoded records.

A computing device comprising:

• a processor;
• and
• a non-transitory storage medium for tangibly storing thereon program logic for execution by the processor, the program logic comprising:
• generating logic executed by the processor for generating, using an unencoded query and an encoding scheme, an encoded query for searching a plurality of records encoded using the encoding scheme, each record of the plurality of records having a plurality of fields,
• the encoding scheme comprising a plurality of encoded value sets corresponding to a plurality of unencoded value sets from the plurality of records, each encoded value set of the plurality of encoded value sets corresponding to a respective one of the plurality of fields,
• an encoded value set, of the plurality of encoded value sets, corresponding to the respective one of the plurality of fields comprising an encoded value for each unencoded value in the unencoded value set corresponding to the respective one of the plurality of fields,
• the encoding scheme being used to encode each of a number of fields, from the plurality of fields, found in the unencoded query, and
• for a respective field of the number of fields found in the unencoded query, the encoding scheme comprising replacing the respective field's unencoded value with a corresponding encoded value from the encoded value set corresponding to the respective field;
• and
• generating logic executed by the processor for generating a search result based on the plurality of records using the encoded query to search the plurality of encoded records.