Dimension reduction using association rules for data mining application

Claim 3

The method of claim 1 above, wherein the step of determining attribute associations further comprises the step of integrating attribute value associations over a value of multiple attributes.

Claim 4

The method of claim 2 above, wherein the step of integrating further comprises the step of integrating attribute value associations that have high support.

Claim 5

The method of claim 1 above, wherein the step of selecting attributes further comprises the step of selecting attributes until a cut-off point is reached.

Claim 6

The method of claim 1 above, further comprising the step of clustering data from the data store using the selected attributes.

Claim 7

The method of claim 1 above, further comprising the step of producing a correlation matrix comprising correlation measures between sets of attributes.

Claim 10

The apparatus of claim 7 above, wherein the means for determining attribute associations further comprises the means for integrating attribute value associations over a value of multiple attributes.

Claim 11

The apparatus of claim 8 above, wherein the means for integrating further comprises the means for integrating attribute value associations that have high support.

Claim 12

The apparatus of claim 7 above, wherein the means for selecting attributes further comprises the means for selecting attributes until a cut-off point is reached.

Claim 13

The apparatus of claim 7 above, further comprising means for clustering data from the data store using the selected attributes.

Claim 14

The apparatus of claim 7 above, further comprising means for producing a correlation matrix comprising correlation measures between sets of attributes.

Claim 15

An article of manufacture comprising a program storage medium readable by a computer and embodying one or more instructions executable by the computer to perform method steps for optimizing data mining, the data mining being performed by the computer to retrieve data from a data store stored on a data storage device coupled to the computer, the data store having records that have multiple attributes, the method comprising the steps of: clustering the attributes to produce sets of attributes, wherein each set has one or more attributes and wherein attributes that have similar correlations with other attributes are grouped into one set of attributes, resulting in non-intersecting sets of attributes; and for each non-intersecting set of attributes, determining attribute value associations between attributes and their values; determining attribute associations from the determined attribute value associations by determining a support value for each attribute value association and determining a cumulative support for each attribute in each attribute value association based on the support values for each attribute value association; and selecting attributes based on the determined attribute associations for performing data mining.

Claim 16

An article of manufacture comprising a program storage medium readable by a computer and embodying one or more instructions executable by the computer to perform method steps for optimizing data mining, the data mining being performed by the computer to retrieve data from a data store stored on a data storage device coupled to the computer, the data store having records that have multiple attributes, the method comprising the steps of: clustering the attributes to produce sets of attributes, wherein each set has one or more attributes and wherein attributes that have similar correlations with other attributes are grouped into one set of attributes, resulting in non-intersecting sets of attributes; and for each non-intersecting set of attributes, determining attribute value associations between attributes and their values; determining attribute associations from the determined attribute value associations by determining a support value for each attribute value association and determining a cumulative support for each attribute in each attribute value association based on the support values for each attribute value association; and selecting attributes based on the determined attribute associations for performing data mining.

Claim 17

The article of manufacture of claim 13 above, wherein the step of determining attribute associations further comprises the step of integrating attribute value associations over a value of multiple attributes.

Claim 18

The article of manufacture of claim 14 above, wherein the step of integrating further comprises the step of integrating attribute value associations that have high support.

Claim 19

The article of manufacture of claim 13 above, wherein the step of selecting attributes further comprises the step of selecting attributes until a cut-off point is reached.

Claim 20

The article of manufacture of claim 13 above, further comprising clustering data from the data store using the selected attributes.

Claim 21

The article of manufacture of claim 13 above, further comprising the step of producing a correlation matrix comprising correlation measures between sets of attributes.