System and method for integrating data mining into a relational database management system
First Claim
1. A computer program product, implemented on a machine readable medium, comprising instructions operable to:
- integrate a relational database with a data mining engine; and
define a data mining model as a relational table, the data mining model providing multiple relationships between inputs and outputs and indicia of the importance of each relationship between an output and one or more inputs.
6 Assignments
0 Petitions
Accused Products
Abstract
An integrated data mining and relational database management system is provided on a server. The data mining engine determines characteristics of relationships between input data values and an output data value that are obtained from a relational database (managed by the relational database management system). The integration allows direct conversion of data values from the relational database to data mining identifiers used for data mining operations by the data mining engine, and vice-versa. The integrated server also makes patterns uncovered during data mining (e.g., “understand” and “predict” information) available in virtual relational database tables that can be queried.
146 Citations
42 Claims
-
1. A computer program product, implemented on a machine readable medium, comprising instructions operable to:
-
integrate a relational database with a data mining engine; and
define a data mining model as a relational table, the data mining model providing multiple relationships between inputs and outputs and indicia of the importance of each relationship between an output and one or more inputs. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A method of evaluating a query on a predict table in a data mining model, the predict table providing multiple relationships between input values and output values and configured to return a predicted output value in response to a query containing one or more input values, the method comprising:
-
(a) identifying an expression or sub-expression in a predicate of the query;
(b) identifying a start/stop condition from the expression or sub-expression;
(c) binding the start/stop condition to a column of the predict table; and
(d) passing column values constrained by the start/stop condition to data mining engine. - View Dependent Claims (29, 30, 31, 32)
identifying a second sub-expression in the predicate; and
determining that the second sub-expression will be evaluated after receiving results of a data mining operation from the data mining engine.
-
-
30. The method of claim 29, wherein the second sub-expression is saved for binding to results of a data mining operation performed by the data mining engine on column values passed to it.
-
31. The method of claim 30, wherein the determining whether the expression or sub-expression is an atomic conjunctive expression or sub-expression, further comprises determining whether the expression or sub-expression compares a bindable value to a model input.
-
32. The method of claim 28, further comprising determining whether the expression or sub-expression is an atomic conjunctive expression or sub-expression, prior to identifying the start/stop condition.
-
33. A computer system comprising:
-
a relational database management system;
a data mining engine integrated with the relational database management system such that data values from a relational database managed by the relational database management system are directly converted to data mining identifiers used for data mining operations by the data mining engine, wherein said data mining engine determines characteristics of relationships between input data values and an output data value and wherein the input and output data values are obtained from the relational database. - View Dependent Claims (34, 35, 36, 37)
a definition of one or more data mining sub-tables which are not populated with data values; - and
one or more data model look up indexes specifying translations between values in the relational database and data mining identifiers used by the data mining engine to represent the relational database values.
-
-
36. The computer system of claim 35, wherein the one or more data model indexes include a forward translation index configured to translate data values from the relational database to data mining identifiers.
-
37. The computer system of claim 35, wherein the one or more data model indexes include a reverse translation index configured to translate data mining identifiers to corresponding data values from the relational database.
-
38. A computer program product, implemented on a machine readable medium, comprising instructions operable to:
-
receive a create model statement at a relational database with data mining capabilities; and
define a data mining model including at least one of a model table, an understand table, a predict table, a binary table, and an index at the relational database with data mining capabilities. - View Dependent Claims (39, 40, 41, 42)
receive an insert statement; and
in response to receiving the insert statement;
populate the data mining model with data; and
provide information about patterns to at least one of the understand table, predict table, and binary table.
-
-
40. The computer program product of claim 38, wherein the instructions are operable to:
-
receive a query of the understand table; and
identify relevant patterns that were uncovered during data mining.
-
-
41. The computer program product of claim 38, wherein the instructions are operable to:
-
receive a query of a predict table; and
return a likely output for a given set of input values.
-
-
42. The computer program product of claim 38, wherein the computer program product is implemented on a machine readable medium at a server and wherein the instructions are operable to at least one of:
-
download the data mining model to a client using the binary table; and
upload a data mining model defined at the client using the binary table.
-
Specification