Extensibility for sales predictor (SPE)

• US 9,953,331 B2
• Filed: 11/28/2016
• Issued: 04/24/2018
• Est. Priority Date: 07/02/2012
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Yaseen, Rahim Mohamed, Shah, Hetal J., Bhatia, Manish, Nagarajan, Chethan Rathna, Manjunath, Jayashree, Mulukutla, Krishna Murty
• Primary Examiner(s): Choi, Peter H
• Assistant Examiner(s): Karmis, Alissa D
• Application Number:
  US15/362,764
• Publication Number:
  US 20170076305A1
• Time in Patent Office: 512 Days
• Field of Search: 705 731, 717113, 707803, 707812, 718100
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Disclosed are methods and systems for implementing extensibility in sales prediction engines. An extensibility framework may be used to modify the metadata schema of the data used by the sales prediction engine to account for extended attributes and entities. The sales prediction engine is also modified to recognize the extended attributes and entities so that a user will be able to create new rules and train new models based on the extended attributes and entities.

Claims

1. A non-transitory computer readable medium having instructions which, when executed by a processor, causes the processor to perform a process for extending attributes for a predictive analysis engine, the process comprising:

defining one or more extended attributes of a database entity,

the database entity corresponding to a database table, the database table comprising one or more original attribute columns and one or more unused extension columns for the database entity, the one or more unused extension columns providing extensibility of the database entity for the one or more extended attributes not defined in the one or more original attribute columns of the database table of the database entity;

receiving from an interface a definition of the one or more extended attributes for the database entity;

modifying a metadata schema of the database table using the one or more extended attributes, comprising;

modifying a first version of the database entity to a second version of the database entity, the second version of the database entity comprising the one or more extended attributes that do not exist in the first version, by;

reviewing the one or more unused extension columns for the first version of the database entity,

identifying unused extension columns from among the one or more unused extension columns of the database table for the first version of the database entity, and

editing the metadata schema for the database table to map the identified unused extension columns from the database table to the one or more extended attributes to generate the second version of the database entity within the database table;

recognizing, by the predictive analysis engine, the one or more extended attributes of the database entity; and

executing the predictive analysis engine to generate recommendations based at least in part on the one or more extended attributes by using one or more new rules or one or more new models, the one or more new rules or the one or more new models comprise the one or more extended attributes.

2. The non-transitory computer readable medium of claim 1, the process further comprising:

identifying the one or more extended attributes to a business intelligence server, wherein a mapping is performed in the business intelligence server to correlate a logical representation of the database entity to the metadata schema of the database table for the second version of the database entity to create database queries to include the one or more extended attributes; and

interfacing with the predictive analysis engine to perform predictive analysis using the one or more extended attributes and data for the one or more extended attributes retrieved with the database queries created in the business intelligence server.

3. The non-transitory computer readable medium of claim 1, wherein the one or more new rules identify characteristics of customers that are more likely to lead to a sale of a particular product.

4. The non-transitory computer readable medium of claim 1, wherein the one or more extended attributes is a province attribute.

5. The non-transitory computer readable medium of claim 1, wherein the one or more extended attributes support extended metrics and multi-tenancy.

6. The non-transitory computer readable medium of claim 1, wherein data for the predictive analysis engine is stored in a transactional system.

7. The non-transitory computer readable medium of claim 1, wherein data for the predictive analysis engine is stored in both a transactional system and a data warehouse.

8. The non-transitory computer readable medium of claim 6, wherein the transactional system comprises an “

out of the box”

schema for various entities.

9. The non-transitory computer readable medium of claim 6, wherein the transactional system stores saved models and rules previously trained or defined by a user.

10. The non-transitory computer readable medium of claim 1, wherein the predictive analysis engine generates reports based at least in part on the one or more extended attributes via view objects.

11. The non-transitory computer readable medium of claim 2, wherein the business intelligence server comprises a repository definition that provides an abstraction layer that maps logical entities and attributes to physical data.

12. The non-transitory computer readable medium of claim 7, wherein models and rules data are stored in the transactional system and entities and attributes data are stored in the data warehouse.

13. The non-transitory computer readable medium of claim 1, the process further comprising generating a rules dictionary that incorporates the one or more extended attributes.

14. The non-transitory computer readable medium of claim 1, the process further comprising training the one or more new models with the one or more extended attributes by:

analyzing historical data associated with the one or more extended attributes, and

applying mathematical models to calculate correlations between products and the one or more extended attributes.

15. The non-transitory computer readable medium of claim 14, wherein the mathematical models comprise a linear regression model to calculate the correlations.

16. The non-transitory computer readable medium of claim 1, wherein a user interface of the predictive analysis engine comprises a select metadata user interface and a rules user interface, the select metadata user interface displaying the one or more extended attributes for selection and training, the rules user interface is used for creating and editing rules.

17. The non-transitory computer readable medium of claim 16, wherein the select metadata user interface selects the one or more extended attributes for learning.

18. The non-transitory computer readable medium of claim 1, wherein the one or more extended attributes for the database entity is associated with a customer entity and wherein the predictive analysis engine is a sales prediction engine.

19. A computer-based method, executed by a processor, for extending attributes for a predictive analysis engine, comprising:

defining one or more extended attributes of a database entity,

the database entity corresponding to a database table, the database table comprising one or more original attribute columns and one or more unused extension columns for the database entity, the one or more unused extension columns providing extensibility of the database entity for the one or more extended attributes not defined in the one or more original attribute columns of the database table of the database entity;

receiving from an interface a definition of the one or more extended attributes for the database entity;

modifying a metadata schema of the database table using the one or more extended attributes, comprising;

modifying a first version of the database entity to a second version of the database entity, the second version of the database entity comprising the one or more extended attributes that do not exist in the first version, by;

reviewing the one or more unused extension columns for the first version of the database entity,

identifying unused extension columns from among the one or more unused extension columns of the database table for the first version of the database entity, and

editing the metadata schema for the database table to map the identified unused extension columns from the database table to the one or more extended attributes to generate the second version of the database entity within the database table;

recognizing, by the predictive analysis engine, the one or more extended attributes of the database entity; and

executing the predictive analysis engine to generate recommendations based at least in part on the one or more extended attributes by using one or more new rules or one or more new models, the one or more new rules or the one or more new models comprise the one or more extended attributes.

20. A computer-based system for extending attributes for a predictive analysis engine, comprising:

a computer processor to execute a set of program code instructions;

a memory to hold the set of program code instructions, in which the set of program code instructions comprises program code to;

define one or more extended attributes of a database entity,

the database entity corresponding to a database table, the database table comprising one or more original attribute columns and one or more unused extension columns for the database entity, the one or more unused extension columns providing extensibility of the database entity for the one or more extended attributes not defined in the one or more original attribute columns of the database table of the database entity;

receiving from an interface a definition of the one or more extended attributes for the database entity;

modify a metadata schema of the database table using the one or more extended attributes, comprising;

modifying a first version of the database entity to a second version of the database entity, the second version of the database entity comprising the one or more extended attributes that do not exist in the first version, by;

reviewing the one or more unused extension columns for the first version of the database entity,

identifying unused extension columns from among the one or more unused extension columns of the database table for the first version of the database entity, and

editing the metadata schema for the database table to map the identified unused extension columns from the database table to the one or more extended attributes to generate the second version of the database entity within the database table;

recognize, by the predictive analysis engine, the one or more extended attributes of the database entity; and

execute the predictive analysis engine to generate recommendations based at least in part on the one or more extended attributes by using one or more new rules or one or more new models, the one or more new rules or the one or more new models comprise the one or more extended attributes.