Method and system for data processing for pattern detection

A program storage device readable by a machine, embodying a program of instructions executable by the machine to perform method steps for analyzing a data source to find association relationships between attributes recorded in the data source, said method steps comprising:

• receiving two or more data records, each record containing one or more of attributes of interest;
• defining at least two attribute k-patterns by reference to a subset of the attributes, where k is an integer and represents the number of mapped attributes that are included in a k-pattern and k is assigned at least two non-zero integer values;
• using an invertible mapping algorithm, mapping each of the at least two attribute k-patterns into an associated key;
• determining a distribution pattern for each of the at least two attribute k-patterns defined in a preceding step;
• inverting the mapping to decode from the keys to the corresponding attribute k-pattern;
• and
• using the keys, associating the distribution patterns determined with each of the at least two attribute k-patterns.

A computer system for analyzing a data source to find association relationships between attributes recorded in the data, comprising:

• at least one processor and accessible memory;
• and
• a computer program accessible for execution by the processor, said program containing the following components:
• (a) instructions for receiving two or more data records, each record containing one or more of attributes of interest;
• (b) instructions for receiving a definition of at least two attribute k-patterns by reference to a subset of the attributes, where k is an integer and represents the number of mapped attributes that are included in the k-pattern and k is assigned at least two non-zero integer values
• (c) instructions for using an invertible mapping algorithm to map each of the at least two attribute k-patterns into an associated key;
• (d) instructions for determining a distribution pattern for each of the at least two attribute k-patterns defined in a preceding step;
• (e) instructions for inverting the mapping to decode from the keys to the associated attribute k-pattern;
• and
• (f) instructions for associating the distribution patterns determined with each of the at least two attribute k-patterns.

A data structure stored in a computer memory, said data structure developed by the following method:

• (a) receiving two or more data records, each record containing one or more of attributes of interest;
• (b) defining at least two attribute k-patterns by reference to a subset of the attributes, where k is an integer and represents the number of attributes that are included in a k-pattern;
• (c) defining an invertible mapping algorithm for mapping each of the at least two attribute k-patterns into an associated key;
• (d) analyzing the data records and determining an actual distribution pattern for records containing each of the at least two attribute k-patterns defined in step (b);
• and
• (e) storing the actual distribution patterns in accordance with the mapping to permit decoding from the keys to the corresponding attribute k-pattern.