Platform independent plug-in methods and systems for data mining and analytics

• US 8,417,715 B1
• Filed: 12/19/2008
• Issued: 04/09/2013
• Est. Priority Date: 12/19/2007
• Status: Active Grant

• Current Assignee: Ramann Kamalakannan, Tilmann Bruckhaus
• Original Assignee: Ramann Kamalakannan, Tilmann Bruckhaus
• Inventors: Bruckhaus, Tilmann, Kamalakannan, Ramann
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Chau, Dung K
• Application Number:
  US12/339,088
• Time in Patent Office: 1,572 Days
• Field of Search: 707/754, 707/776, 707/999.004, 707/999.006, 707/999.102, 709/248
• US Class Current: 707/758
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

The present invention comprises methods, system, and apparatuses for generating and delivering analytic results for any simple or highly complex problem for which data exists that software or similar automated means can analyze. The present invention thus contemplates methods, systems, apparatuses, software, software processes, computer-readable medium, and/or data structures to enable performance of these and other features. In one embodiment, a method of the present invention comprises extracting and converting data using a data management component into a form usable by a data mining component, performing data mining to develop a model in response to a question or problem posed by a user.

Claims

1. A system for providing a result based on a task, comprising:

a first module comprising a first non-transitory computer-readable memory storing computer-executable code that receives a business task related to a business of a user;

a second module comprising a second non-transitory computer-readable memory storing computer-executable code that prepares the business task received by the first module for processing a machine-executable task by retrieving a pre-configured mapping between the business task and the machine-executable task stored in a business task repository;

wherein the second module further comprises;

a first sub-module comprising a first sub-module non-transitory computer-readable memory storing computer-executable code that retrieves the pre-configured mapping between the business task and the machine-executable task;

a second sub-module comprising a second sub-module non-transitory computer-readable memory storing computer-executable code that specifies a source table, or a plurality of source tables and a plurality of data fields based on the pre-configured mapping associated with the definition of the machine-executable task;

a third sub-module comprising a third sub-module non-transitory computer-readable memory storing computer-executable code that places the selected data into a table for later use and preserving a longitudinal state of a plurality of selected data;

a third module comprising a third non-transitory computer-readable memory storing computer-executable code that selects data from a database to produce selected data, wherein the data is selected based on the pre-configured mapping, wherein the pre-configured mapping is associated with a definition of the machine-executable task;

a fourth module comprising a fourth non-transitory computer-readable memory storing computer-executable code that analyzes the selected data to produce analyzed data by selecting and applying at least one data mining algorithm from an algorithm library based on the pre-configured mapping associated with the definition of the machine-executable task;

a fifth module comprising a fifth non-transitory computer-readable memory storing computer-executable code that determines at least one result based on the analyzed data and the at least one data mining algorithm based on the pre-configured mapping associated with the definition of the machine-executable task; and

a sixth module comprising a sixth non-transitory computer-readable memory storing computer-executable code that sends the at least one result for display based on the pre-configured mapping associated with the definition of the machine-executable task.

2. The system of claim 1, further comprising;

a seventh module comprising a seventh non-transitory computer-readable memory storing computer-executable code that executes the at least one data mining algorithm to generate a first model, wherein the first model is based on the selected data, and the at least one result is based on the first model.

3. The system of claim 2, wherein the seventh module further comprises computer-executable code that executes the at least one data mining algorithm to generate a second model, wherein the second model is based on the selected data, and wherein the system further comprises:

an eighth module comprising an eighth non-transitory computer-readable memory storing computer-executable code that selects a best model from the first model and the second model, and wherein the at least one result is based on the best model.

4. The system of claim 3, wherein the seventh module further comprises computer-executable code that executes a plurality of data mining algorithms to generate a plurality of models and wherein the eighth module further comprises computer-executable code that selects the best model from the plurality of models.

5. The system of claim 3, further comprising;

a ninth module comprising a ninth non-transitory computer-readable memory storing computer-executable code that configures a plurality of new business tasks and mapping the plurality of new business tasks to a plurality of machine-executable tasks.

6. The system of claim 1, wherein the first non-transitory computer-readable memory storing computer-executable code further comprising computer-executable code that receives a plurality of pre-configured business tasks from the business task repository.

7. The system of claim 1, wherein the business task related to a software application.

8. A computer-implemented method for providing a result based on a query, comprising:

receiving, by a first computing subsystem, a business task, wherein the business task is related to a business of a user;

preparing, by a second computing subsystem, the business task received by the first computing subsystem for processing a machine-executable task, by the second computing subsystem, by retrieving a pre-configured mapping between the business task and the machine-executable task in a business task repository;

wherein the second computing subsystem further comprises;

retrieving, by a first sub-module, the pre-configured mapping between the business task and the machine-executable task;

specifying, by a second sub-module, a source table, or a plurality of source tables and a plurality of data fields based on the pre-configured mapping associated with the definition of the machine-executable task;

placing, by a third sub-module, the selected data into a table for later use and preserving a longitudinal state of a plurality of selected data

selecting, by a third computing subsystem, data from a database to produce selected data, wherein the data is selected based on the pre-configured mapping, wherein the pre-configured mapping is associated with a definition of the machine-executable task;

analyzing, by a fourth computing subsystem, the selected data to produce analyzed data and selecting, by the fourth computing subsystem, at least one data mining algorithm from an algorithm library to produce a selected algorithm, wherein the selecting the at least one data mining algorithm is based on the pre-configured mapping associated with the definition of the machine-executable task;

determining, by a fifth computing subsystem, at least one result based on the analyzed data and the selected algorithm based on the pre-configured mapping associated with the definition of the machine-executable task;

sending, by a sixth computing subsystem, the at least one result for display based on the pre-configured mapping associated with the definition of the machine-executable task.

9. The method of claim 8, further comprising;

executing, by a seventh computing subsystem, the at least one data mining algorithm to generate a first model, wherein the first model is based on the selected data and the at least one result is based on the first model.

10. The method of claim 9, further comprising:

executing, by the seventh computing subsystem, the at least one data mining algorithm to generate a second model, wherein the second model is based on the selected data; and

selecting, by an eighth computing subsystem, a best model from the first model and the second model, and wherein the at least one result is based on the best model.

11. The method of claim 10, wherein the executing further comprises executing a plurality of data mining algorithms to generate a plurality of models and wherein the selecting further comprises selecting the best model from the first model, the second model, or one of the plurality of models.

12. The method of claim 10, further comprising:

configuring, by a ninth computing subsystem, a plurality of new business tasks; and

mapping, by the ninth computing subsystem, the plurality of new business tasks to a plurality machine-executable tasks.

13. The method of claim 8, further comprising:

retrieving, by a first module the pre-configured mapping between the business task and the machine-executable;

specifying, by a second module, a source table, or a plurality of source tables and a plurality of data fields based on the pre-configured mapping associated with the definition of the machine-executable task;

placing, by a third module, the selected data into a table for later use and preserving longitudinal state of a plurality of selected data.

14. The method of claim 8, wherein the receiving by the first computing subsystem further comprises receiving a plurality of preconfigured business tasks from the business task repository.

15. The method of claim 8, wherein the business task relates to a software application.

16. A system for preparing a task for processing,

comprising:

a first module comprising a non-transitory computer-readable memory storing computer-executable code that retrieves a pre-configured mapping associated with a business task, wherein the pre-configured mapping contains mapping between the business task and a machine-executable task stored in a business task repository;

wherein the pre-configured mapping associated with the business task further includes an identity of the business task, wherein the business task repository further contains a plurality of pre-configured mapping to identify a plurality of business tasks, and wherein any one or more of the plurality of business tasks is associated with a plurality of machine-executable tasks;

a second module comprising a second non-transitory computer-readable memory storing computer-executable code that specifies a source table, or a plurality of source tables and a plurality of data fields from a first database on the pre-configured mapping, wherein the pre-configured mapping is associated with a definition of the machine-executable task;

a third module comprising a third non-transitory computer-readable memory story computer-executable code that selects data from the source table or the plurality of source tables to produce selected data, wherein the data is selected based on the pre-configured mapping associated with the definition of the machine-executable task;

a fourth module comprising a fourth non-transitory computer-readable memory storing computer-executable code that places the selected data into a first table for later use and preserving a longitudinal state of selected data and a second database for storing the first table; and

a fifth module comprising a fifth non-transitory computer-readable memory storing computer-executable code that selects and applies at least one data mining algorithm from an algorithm library, wherein the selecting the at least one data mining algorithm is based on the pre-configured mapping associated with the definition of the machine-executable task.

17. The system of claim 16, wherein the fourth module further comprises computer-executable code that places the selected data into the first table based on a pre-defined format based upon a pre-determined platform used to create the first database.

18. The system of claim 16, wherein the fourth non-transitory computer-readable memory storing computer-executable code further comprises:

computer-executable code that applies mathematical and statistical functions to the data in the first table to derive additional inputs for the data mining algorithm;

computer-executable code that stores additional inputs in a second table in the second database for a later use by the data mining algorithm; and

computer-executable code that transforms the data stored in the first and second tables in the second database into a de-normalized table that allows for usage of data regardless of form by the data mining algorithm.

19. A computer-implemented method for preparing a task for processing, comprising;

retrieving, by a first computing subsystem, a pre-configured mapping associated with a business task, wherein the pre-configured mapping comprises a mapping between the business task and a machine-executable task stored in a business task repository;

wherein the pre-configured mapping associated with the business task further includes an identity of the business task, wherein the business task repository further contains a plurality of pre-configured mapping to identify a plurality of business tasks, and wherein any one or more of the plurality of business tasks is associated with a plurality of machine-executable tasks;

specifying, by a second computing subsystem, a source table, or a plurality of source tables and a plurality of data fields from a first database based on the pre-configured mapping, wherein the pre-configured mapping is associated with a definition of the machine-executable task;

selecting, by a third computing subsystem, data from the source table or the plurality of source tables to produce selected data, wherein the data is selected based on the pre-configured mapping associated with the definition of the machine-executable task;

placing, by a fourth computing subsystem, the selected data into a first table for later use and preserving a longitudinal state of the selected data and storing the first table in a second database; and

selecting and applying at least one data mining algorithm from an algorithm library, wherein the selecting the at least one data mining algorithm is based on the pre-configured mapping associated with the definition of the machine-executable task.

20. The method of claim 19, wherein the placing the selected data into the first table is based on a pre-defined format based upon a pre-determined platform used to create the first database.

21. The method of claim 19, further comprising:

deriving, by the fourth computing subsystem, additional inputs for the data mining algorithm by applying mathematical and statistical functions to the data in the first table;

storing, by the fourth computing subsystem, of the additional inputs in a second table in the second database for a later use by the data mining algorithm; and

transforming, by the fourth computing subsystem, of the data stored in the first and second tables in the second database into a de-normalized table that allows for usage of data regardless of form by the data mining algorithm.