STATISTICALLY AND ONTOLOGICALLY CORRELATED ANALYTICS FOR BUSINESS INTELLIGENCE

• US 20160232464A1
• Filed: 03/28/2016
• Published: 08/11/2016
• Est. Priority Date: 02/11/2015
• Status: Active Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Shen-Hsieh, Angela B., Nonez, Robert Y.
• Application Number:
  US15/082,352
• Publication Number:
  US 20160232464A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06Q 10/0633 Workflow analysis
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Techniques are disclosed for statistically and ontologically correlated business intelligence (BI) analytics. An example method includes performing an ontological analysis on relevant data defined for a BI analytics query to determine correlations with ontological concepts. The method includes performing a statistical analysis on direct analytics output data to rank the direct analytics output data in order of influence on the direct analytics output. The method includes performing a statistical analysis on the relevant data set relative to the direct analytics output data to determine data in the relevant data set that influence the direct analytics output data, thereby generating a list of key drivers ranked in order of influence. The method includes revising the ranking of the key drivers based on correlations of the key drivers with the ontological concepts. The method includes generating a correlated analytics output comprising information on the key drivers based on the ranking.

Claims

1. A method for business intelligence (BI) analytics, the method comprising:

performing, by one or more processing devices, an ontological analysis on data items in a relevant data set defined for a BI analytics query to determine one or more correlations of the data items in the relevant data set with ontological concepts in an ontological concept subsystem;

performing, by the one or more processing devices, a first statistical analysis on a set of direct analytics output data items from the relevant data set that are included in a direct BI analytics output to rank the direct analytics output data items in an order of influence on the direct BI analytics output;

performing, by the one or more processing devices, a second statistical analysis on the data items in the relevant data set relative to the direct analytics output data items to determine one or more of the data items in the relevant data set that influence the respective direct analytics output data items, thereby generating a list of key drivers from among the data items in the relevant data set such that the list of key drivers has a ranking in an order of the influence;

revising, by the one or more processing devices, the ranking of the list of key drivers based at least in part on the correlations of the key drivers with the ontological concepts; and

generating, by the one or more processing devices, a correlated analytics output comprising information on one or more of the key drivers based on the ranking of the list of key drivers.

2. The method of claim 1, further comprising removing, from the list of key drivers, key drivers that are already present in the direct analytics output data items.

3. The method of claim 1, wherein generating the correlated analytics output comprising the information on the one or more of the key drivers based on the ranking of the list of key drivers further comprises generating the correlated analytics output with information on how the key drivers influence the direct analytics output data items.

4. The method of claim 1, wherein generating the correlated analytics output comprising the information on the one or more of the key drivers comprises generating the correlated analytics output comprising correlations between one or more of the key drivers and one or more of the direct analytics output data items based on the ranking of the list of key drivers.

5. The method of claim 1, further comprising selecting a visualization for the correlated analytics output based on the key drivers.

6. The method of claim 1, further comprising:

determining statistical information about one or more of the key drivers comprising one or more of a minimum, a maximum, and an average value of the one or more of the key drivers,

wherein generating the correlated analytics output further comprises including the one or more of the minimum, the maximum, and the average value of the one or more of the key drivers in the correlated analytics output.

7. The method of claim 1, wherein revising the ranking of the list of key drivers based at least in part on the correlations of the key drivers with the ontological concepts comprises ranking key drivers that have a correlation with one of the ontological concepts higher than key drivers that do not have a correlation with one of the ontological concepts.

8. The method of claim 1, wherein revising the ranking of the list of key drivers based at least in part on the correlations of the key drivers with the ontological concepts comprises removing, from the list of key drivers, key drivers that do not have a correlation with one of the ontological concepts.

9. The method of claim 1, further comprising providing drill-down functionality in the correlated analytics output to enable access to additional data sources related to the information in the correlated analytics output.

10. The method of claim 1, further comprising:

receiving a new input based on the correlated analytics output; and

generating a subsequent output comprising additional information related to the key drivers and responsive to the new input.