Data aggregation user interface and analytic adapted for a KStore

• US 7,389,301 B1
• Filed: 06/10/2005
• Issued: 06/17/2008
• Est. Priority Date: 06/10/2005
• Status: Active Grant

• Current Assignee: Unisys Corporation
• Original Assignee: Unisys Corporation
• Inventors: Mazzagatti, Jane Campbell, Rajcan, Steven L., Buckwalter, Robert
• Primary Examiner(s): Alam; Shahid
• Assistant Examiner(s): NGUYEN, PHONG H
• Application Number:
  US11/150,663
• Time in Patent Office: 1,103 Days
• Field of Search: 707/3, 707/101, 707/102, 707/103.R, 707/104.1
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2246 Trees, e.g. B+trees
• CPC Class Codes: G06F 16/24556 Aggregation; Duplicate elimination
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

Aggregation of data in an interlocking trees datastore, especially when the interlocking datastore is a KStore is described. It details consolidating data into a summary or aggregation so that some particular desired analytic type of operation may easily be performed on the data. It uses a set of data constraints across the entire data set. This redefines the data set, which may be for example, individual receipts granular by week or month. When data is learned into a KStore, aggregation parameters may be collected and these parameters may be used to constrain the dataset recorded in K, and direct performance of an analytic on a particular a field value(s). Additional features and details are provided within.

Claims

1. A computer implemented method for obtaining desired information from an interlocking trees datastore comprising:

establishing aggregation parameters from received input,

traversing said interlocking trees datastore comprising elemental root nodes, subcomponent nodes and end-product nodes using asCase and asResult bidirectional links for determining paths within the interlocking trees datastore structure, with said established aggregation parameters so as to establish a set of paths within said interlocking trees datastore in accord with said established aggregation parameters,

wherein said interlocking trees data store has already recorded a collection of data such that relationships of said collection of data are modeled by said interlocking trees datastore, said interlocking trees datastore situated within memory accessible to a process running in a computer, said interlocking trees datastore comprising a structured collection of nodes connected by links of said nodes having pointers to other nodes of said interlocking trees datastore, wherein said nodes containing a count field, said nodes including at least nominally different kinds of nodes, a first kind called root nodes of which there are at least one primary root node and at least one elemental root node and which may include other root nodes, a second kind of node called an end of thought (EOT) node, at least one node of a third kind of node called a subcomponent node, and at least one node of a kind of node called an end product node, and wherein there exist at least two kinds of said links, Result and Case links, wherein said Result links point between a one of said root nodes and any other node, and wherein said bi-directional Result links point between a Result field of one node and an asResult list of another node, and wherein said bi-directional Case links point between the Case field of one node and an asCase list of another node.

2. A computer implemented method as set forth in claim 1 for obtaining desired information from a collection of data further comprising:

building an interlocking trees datastore via a Learn Engine reading data from a selected data set, and recording the relationships within said data set into said interlocking trees datastore.

3. The method of claim 1 further comprising, determining an analytic to apply to said set of K paths of said dataset, and, applying said analytic to said established set of K paths in said interlocking trees datastore to produce results.

4. The method of claim 3 wherein the results are in the form of an aggregated data set.

5. The method of claim 4 wherein the results are produced in the form of a set of summary records.

6. The method of claim 5 wherein the set of summary records is produced in an XML format.

7. The method of claim 3 wherein the results are a set of K paths.

8. The method of claim 3 further comprising displaying said results.

9. The method of claim 8 wherein said displaying of said results is accomplished via a printout or a visual display screen having displayed thereon at least a report, chart, graph, grouping spreadsheet or other visually perceptible image related to the results.

10. The method of claim 3 wherein said analytic is applied only to specific fields of nodes in said established set of K paths in said interlocking trees datastore.

11. The method of claim 3 wherein said established set of K paths contain nodes that contain at least count fields.

12. The method of claim 11 wherein said analytic is applied to the value of the counts in the count fields.

13. The method of claim 10 wherein at least one of said specific fields is a variable field, and wherein values of said variable are used by said analytic to produce said results.

14. The method of claim 13 wherein obtaining said values of said variable is accomplished through

identifying any field variables needed to satisfy the fields,

traversing said interlocking trees datastore to root nodes having values for said identified any field variables, and

constructing variables from variable fields in said root nodes having said values.

15. The method of claim 14 wherein said constructed variables are summed.

16. The method of claim 3 wherein said analytic is at least one of a set of selectable analytics from the group comprising:

sum, average, distinct count, distinct count percentage, distinct count ratio, record count, record count percentage, record count ratio, probability, Bayesian Classification analysis, decision tree, inultivariate analysis, association rules, market basket.

17. The method of claim 3 wherein the analytic is a form of analytic commonly used for Business Intelligence, Data Mining, or statistical data analysis projects.

18. A computer system having software and hardware therein which when operating in said computer system:

establishes aggregation parameters from received input,

traverses an interlocking trees datastore comprising elemental root nodes, subcomponent nodes and end-product nodes using asCase and asResult bidirectional links for determining paths within the interlocking trees datastore structure, with said established aggregation parameters so as to establish a set of paths within said interlocking trees datastore in accord with said established aggregation parameters,

wherein said interlocking trees data store has already recorded a collection of data such that relationships of said collection of data are modeled by said interlocking trees datastore, said interlocking trees datastore situated within memory accessible to a process running in a computer, said interlocking trees datastore comprising a structured collection of nodes connected by links of said nodes having pointers to other nodes of said interlocking trees datastore, wherein said nodes containing a count field, said nodes including at least nominally different kinds of nodes, a first kind called root nodes of which there are at least one primary root node and at least one elemental root node and which may include other root nodes, a second kind of node called an end of thought (EOT) node, at least one node of a third kind of node called a subcomponent node, and at least one node of a kind of node called an end product node, and wherein there exist at least two kinds of said links, Result and Case links, wherein said Result links point between a one of said root nodes and any other node, and wherein said bi-directional Result links point between a Result field of one node and an asResult list of another node, and wherein said bi-directional Case links point between the Case field of one node and an asCase list of another node.