US 7,493,296 B2
Method and system for classifying information
Rick A. Hamilton, II, Charlottesville, Va. (US); Gino A. Palozzi, Westerville, Ohio (US); Evan A. Salop, Riverside, Conn. (US); and Timothy M. Waters, Hiram, Ga. (US)
Assigned to International Business Machines Corporation, Armonk, N.Y. (US)
Filed on May 31, 2006, as Appl. No. 11/421,216.
Prior Publication US 2007/0282897 A1, Dec. 06, 2007
Int. Cl. G06F 17/00 (2006.01); G06F 12/00 (2006.01); G06F 17/30 (2006.01); G06N 5/00 (2006.01)
U.S. Cl. 706—45  [707/200] 1 Claim
OG exemplary drawing
 
1. A computer-implemented method for classifying information comprising:
determining a plurality of management characteristics for an information type;
determining a plurality of business characteristics for the information type;
determining at least one management tally for the information type based on the management characteristics;
determining a management characteristic scoring scale;
assigning a management characteristic score to each management characteristic for the information type;
calculating a linear combination of the management characteristic scores associated with the information type;
determining at least one business tally for the information type based on the business characteristics wherein determining the business tally comprises:
determining a business characteristic scoring scale;
assigning a business characteristic score to each business characteristic for the information type; and
calculating a linear combination of the business characteristic scores associated with the information type;
determining a normalization factor for the business tally;
determining at least one management weighting factor for each or for all of the management characteristics;
determining at least one business weighting factor for each or for all of the business characteristics;
determining an information class gradient based on the management tally and an information class value wherein the information class gradient includes a plurality of information classes having an information class range;
providing the information class value wherein the information class value represents the number of information classes in the information class gradient;
calculating a maximum possible management tally wherein the maximum possible management tally is a summation of a maximum management characteristic score for each management characteristic;
calculating the information class range as the quotient of the maximum possible management tally and the information class value;
determining at least one management information class based on the management tally and the information class gradient;
determining at least one business information class based on the business tally and the information class gradient;
determining a global information class for the information type based on the management tally and the business tally wherein determining the global information class for the information type comprises:
combining the at least one management weighting factor with the management tally to yield a weighted management tally;
combining the at least one business weighting factor with the business tally to yield a weighted business tally;
calculating a global tally based on the weighted management tally and the weighted business tally;
mapping the global tally into the information class gradient, and continuously updating the global tally based on an evaluation of the global information class;
defining the management characteristics;
defining the business characteristics; and
storing the information type based on the information class gradient.