On-line mining of quantitative association rules

Claim 2

A computer program device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for generating quantitative association rules, the method steps comprising: a) receiving a query including antecedent and consequent attributes where said antecedent and consequent attributes further comprise a plurality of quantitative and categorical attributes; b) organizing a relationship between said antecedent and consequent attributes; c) prestoring data defining the relationship between said antecedent attributes and data related to said consequent attributes; and d) deriving one or more quantitative association rules from prestored data in response to said prestoring step.

Claim 3

The computer program device of claim 1, wherein step b further comprises partitioning said antecedent data hierarchically into an index tree where said index tree comprises a multiplicity of index nodes.

Claim 4

The method of claim 2, wherein the step of partitioning said antecedent data hierarchically into the index tree further comprises: a) storing a first value at each index node of said index tree representing the actual support; and b) storing a second value at each index node of said index tree representing the frequency of occurrence for each user query consequent attribute.

Claim 5

The computer program device of claim 1, said answer further comprises one or more quantitative association rules, an actual confidence value associated with each rule, an actual support value associated with each rule, and an interest level associated with each rule.

Claim 6

The computer program device of claim 4, wherein said quantitative association rules consist of only those rules which are interesting, where a interesting rules include those rules whose computed interest level is at least equal to said user defined interest level.

Claim 7

The computer program device of claim 5, wherein said interest level is defined as the minimum of a first and a second computed ratio, wherein said first ratio is defined as the actual confidence divided by an expected confidence and a second ratio is defined as the actual support divided by an expected support, wherein said expected confidence and support are computed values based on a presumption of statistical independence.

Claim 8

The computer program device of claim 1, wherein said antecedent attributes are further comprised of categorical and quantitative attributes.

Claim 9

The computer program device of claim 7, wherein said quantitative attributes are further defined by a range consisting of a lower and upper bound.

Claim 10

The method of claim 1, wherein step d further comprises the steps of: i) searching all index nodes of said index tree to isolate those nodes whose antecedent attribute range corresponds to said user query antecedent attribute range; ii) selecting from those nodes which satisfy the criteria of step i, whose consequent attribute is at least equal to said user defined value of minimum confidence; and iii) building a merge tree from those nodes which satisfy the criteria of steps i and ii.

Claim 11

The computer program device of claim 9, wherein step iii further comprises deleting meaningless nodes and combining other nodes to create said merge tree.

Claim 12

The computer program device of claim 10, wherein a meaningless node is a node which does not have a corresponding calculated value of confidence at least equal to said user defined value of minimum confidence.

Claim 13

The computer program device of claim 10, wherein the merge tree may be built either for a single or multiple consequent attributes.

Claim 14

The computer program device according to claim 1 wherein the step of constructing a merged rule tree comprises a) traversing each node of the unmerged rule tree in post order; b) evaluating each traversed node for inclusion or exclusion in the unmerged rule tree, further comprising the steps of: i) determining whether each said user defined consequent attribute value is greater than the consequent attribute value stored at said node; ii) preserving said node in said merged rule tree when the condition of step i is satisfied; iii) deleting said node from said merged rule tree when the condition of step i is not satisfied and said node has no associated child nodes; iv) deleting said node from said merged rule tree when the condition of step I is not satisfied and said node has one child node v) adjusting the range of said consequent attribute when the condition of step i is not satisfied; vi) directly associating an ancestor node and child node of said deleted node when the condition of step iv is satisfied; and vii) repeating steps i-vi until all nodes have been traversed in post order.

Claim 15

A computer program device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for generating quantitative association rules, the method steps comprising: a) receiving data including a user defined value of minimum support, a user defined value of minimum confidence, and a user query comprising an antecedent and consequent condition where said antecedent and consequent condition further comprise a plurality of quantitative and categorical attributes; b) constructing in memory an index tree comprised of one or more dimensions, where each dimension is defined by one of the quantitative attributes, said index tree including a plurality of index nodes where said index nodes further include a plurality of data records; c) constructing in memory an unmerged rule tree from said index tree; d) constructing in memory a merged rule tree from said unmerged rule tree; e) generating one or more quantitative association rules from those index nodes that satisfy said user query and whose support is at least equal to said minimum support, and whose confidence is at least equal to said minimum confidence; and f) displaying to a user output data including: said quantitative association rules from the generating step; a value of actual confidence associated with each generated quantitative association rule; a value of support associated with each generated quantitative association rule; and a value of interest level associated with each generated quantitative association rule.

Claim 16

A computer program device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for generating quantitative association rules, the method steps comprising: a) receiving data including a user defined value of minimum support, a user defined value of minimum confidence, and a user query comprising an antecedent and consequent condition where said antecedent and consequent condition further comprise a plurality of quantitative and categorical attributes; b) constructing in memory an index tree comprised of one or more dimensions, where each dimension is defined by one of the quantitative attributes, said index tree including a plurality of index nodes where said index nodes further include a plurality of data records; c) constructing in memory an unmerged rule tree from said index tree; d) constructing in memory a merged rule tree from said unmerged rule tree; e) generating one or more quantitative association rules from those index nodes that satisfy said user query and whose support is at least equal to said minimum support, and whose confidence is at least equal to said minimum confidence; and f) displaying to a user output data including: said quantitative association rules from the generating step; a value of actual confidence associated with each generated quantitative association rule; a value of support associated with each generated quantitative association rule; and a value of interest level associated with each generated quantitative association rule.

Claim 17

The computer program device according to claim 14 wherein the step of generating quantitative association rules is repeated so that said user query is interactively modified to further define said association rules.

Claim 18

The computer program device of claim 14 wherein the step of constructing an index tree comprises the steps of: 1) constructing a binary index tree of one or more dimensions, where each dimension is defined by one of said user supplied quantitative antecedent attributes; and 2) storing at each index node said support level and confidence level.

Claim 19

The computer program device of claim 14 wherein the step of constructing an unmerged rule tree comprises the steps of: i) searching each node of said index tree; and ii) selecting those nodes which contain rules which satisfy the user specified consequent condition and have confidence at least equal to said user defined value of minimum confidence, and a value of support at least equal to said user defined value of minimum support.

Claim 20

The computer program device according to claim 17, wherein step ii further comprises: i) constructing a pointer; ii) equating said pointer to a root node in said index tree; iii) adding said node associated with said pointer to a list; iv) adding all children of the node pointed to by said pointer with antecedent attribute wholly contained within the parameters of said user specified antecedent attribute and have a minimum support value at least equal to said user defined minimum support; v) determining whether the data records stored at the node pointed to by said pointer at least equal to the user specified consequent condition and have a confidence at least equal to said user defined minimum confidence for the node pointed by said pointer; vi) generating a quantitative association rule associated with said consequent conditions; vii) deleting said node from said list when the conditions of the previous step are not satisified; viii) determining whether said list is empty; ix) terminating when said list is empty; x) when the condition of step ix is not satisified, equating said pointer to the next node of said index tree; and xi) repeating steps iii-x when the condition of step ix is not satisfied.

Claim 23

The method of claim 19, wherein step b) further comprises partitioning said antecedent data hierarchically into an index tree where said index tree comprises a multiplicity of index nodes.

Claim 24

The method of claim 20, wherein the step of partitioning said antecedent data hierarchically into the index tree further comprises: a) storing a first value at each index node of said index tree representing the actual support; and b) storing a second value at each index node of said index tree representing the frequency of occurrence for each user query consequent attribute.

Claim 25

The method of claim 19 wherein step a) further includes receiving one or more of a user defined value of minimum confidence, a user defined value of minimum support, and a user defined value of interest level.

Claim 26

The method of claim 22 wherein step d) includes deriving one or more quantitative association rules that satisfy said user query, including said one or more of said minimum support, said minimum confidence and said interest level.

Claim 27

The method of claim 23, further including displaying to a user one or more quantitative association rules, an actual confidence value associated with each rule, an actual support value associated with each rule, and an interest level associated with each rule.

Claim 28

The method of claim 24, wherein said quantitative association rules include only those rules which are interesting, where said interesting rules include those rules whose computed interest level is at least equal to said user defined interest level.

Claim 29

The method of claim 25, wherein said interest level is defined as the minimum of a first and a second computed ratio, wherein said first ratio is defined as the actual confidence divided by an expected confidence and a second ratio is defined as the actual support divided by an expected support, wherein said expected confidence and support are computed values based on a presumption of statistical independence.

Claim 30

The method of claim 19, wherein said antecedent attributes are further comprised of categorical and quantitative attributes.

Claim 31

The method of claim 27, wherein said quantitative attributes are further defined by a range consisting of a lower and upper bound.

Claim 32

The method of claim 19, wherein step d) further comprises the steps of: i) searching all index nodes of said index tree to isolate those nodes whose antecedent attribute range corresponds to said user query antecedent attribute range; ii) selecting from those nodes which satisfy the criteria of step i, whose consequent attribute is at least equal to said user defined value of minimum confidence; and iii) building the merge tree from those nodes which satisfy the criteria of steps i and ii.

Claim 33

The method of claim 29, wherein step iii further comprises deleting meaningless nodes and combining other nodes to create said merge tree.

Claim 34

The method of claim 30, wherein a meaningless node is a node which does not have a corresponding calculated value of confidence at least equal to said user defined value of minimum confidence.

Claim 35

The method of claim 30, wherein the merge tree may be built either for a single or multiple consequent attributes.

Claim 36

The method according to claim 19 wherein the step of constructing a merged rule tree comprises a) traversing each node of the unmerged rule tree in post order; b) evaluating each traversed node for inclusion or exclusion in the unmerged rule tree, further comprising the steps of: i) determining whether each said user defined consequent attribute value is greater than the consequent attribute value stored at said node; ii) preserving said node in said merged rule tree when the condition of step i is satisfied; iii) deleting said node from said merged rule tree when the condition of step i is not satisfied and said node has no associated child nodes; iv) deleting said node from said merged rule tree when the condition of step I is not satisfied and said node has one child node v) adjusting the range of said consequent attribute when the condition of step i is not satisfied; vi) directly associating an ancestor node and child node of said deleted node when the condition of step iv is satisfied; and vii) repeating steps i-vi until all nodes have been traversed in post order.

Claim 37

A computer process of online mining for a large database having a plurality of records, each record having a plurality of quantitative and categorical items for providing quantitative association rules comprising the steps of: a) receiving data including a user defined value of minimum support, a user defined value of minimum confidence, a user defined value of interest, and a user query comprising an antecedent and consequent condition where said antecedent and consequent condition further comprise a plurality of quantitative and categorical attributes; b) constructing in memory an index tree comprised of one or more dimensions, where each dimension is defined by one of the quantitative attributes, said index tree including a plurality of index nodes where said index nodes further include a plurality of data records; c) constructing in memory an unmerged rule tree from said index tree; d) constructing in memory a merged rule tree from said unmerged rule tree; e) generating one or more quantitative association rules from those index nodes that satisfy said user query and whose support is at least equal to said minimum support, and whose confidence is at least equal to said minimum confidence; and f) displaying to a user output data including: said quantitative association rules from the generating step; a value of actual confidence associated with each generated quantitative association rule; a value of support associated with each generated quantitative association rule; and a value of interest level associated with each generated quantitative association rule.

Claim 38

A computer process of online mining for a large database having a plurality of records, each record having a plurality of quantitative and categorical items for providing quantitative association rules comprising the steps of: a) receiving data including a user defined value of minimum support, a user defined value of minimum confidence, a user defined value of interest, and a user query comprising an antecedent and consequent condition where said antecedent and consequent condition further comprise a plurality of quantitative and categorical attributes; b) constructing in memory an index tree comprised of one or more dimensions, where each dimension is defined by one of the quantitative attributes, said index tree including a plurality of index nodes where said index nodes further include a plurality of data records; c) constructing in memory an unmerged rule tree from said index tree; d) constructing in memory a merged rule tree from said unmerged rule tree; e) generating one or more quantitative association rules from those index nodes that satisfy said user query and whose support is at least equal to said minimum support, and whose confidence is at least equal to said minimum confidence; and f) displaying to a user output data including: said quantitative association rules from the generating step; a value of actual confidence associated with each generated quantitative association rule; a value of support associated with each generated quantitative association rule; and a value of interest level associated with each generated quantitative association rule.

Claim 39

The method according to claim 34 wherein the step of generating quantitative association rules is repeated so that said user query is interactively modified to further define said association rules.

Claim 40

The method according to claim 34 wherein the step of constructing an index tree comprises the steps of: 1) constructing a binary index tree of one or dimensions, where each dimension is defined by one of said user supplied quantitative antecedent attributes; 2) storing at each index node said support level and confidence level.

Claim 41

The method according to claim 34 wherein the step of constructing an unmerged rule tree comprises the steps of: i) searching each node of said index tree; ii) selecting those nodes which contain rules which satisfy the user specified consequent condition and have confidence at least equal to said user defined value of minimum confidence, and a value of support at least equal to said user defined value of minimum support.

Claim 42

The method according to claim 37, wherein step ii further comprises: i) constructing a pointer; ii) equating said pointer to a root node in said index tree; iii) adding said node associated with said pointer to a list; iv) adding all children of the node pointed to by said pointer with antecedent attribute wholly contained within the parameters of said user specified antecedent attribute and have a minimum support value at least equal to said user defined minimum support; v) determining whether the data records stored at the node pointed to by said pointer at least equal to the user specified consequent condition and have a confidence at least equal to said user defined minimum confidence for the node pointed by said pointer; vi) generating a quantitative association rule associated with said consequent conditions; vii) deleting said node from said list when the conditions of the previous step are not satisfied; viii) determining whether said list is empty; ix) terminating when said list is empty; x) when the condition of step ix is not satisfied, equating said pointer to the next node of said index tree; and xi) repeating steps iii-x when the condition of step ix is not satisfied.