Method and apparatus for automatically providing expert analysis-based advice
The present invention is directed to a computer-based intelligent adviser having particular applicability in industry situations and job positions in which analytical skills are secondary to a position but in which analytical skills are important to achieving the highest level of performance in the business. The present invention contemplates both the process of developing a computer based intelligent adviser as well as a system and process of using a computer-based intelligent adviser.
1. A computer implemented knowledge-based system for providing advisories to a user on the operation of a business, the system comprising:
a database of inference rules relevant to the business;
a source of input data about the business operations;
an output for rendering advisories for the user; and
a processor programmed to:
analyze predicates of the inference rules to identify data to be processed and operations to be performed prior to applying the inference rules;
process the input data to calculate a plurality of aggregate values based on the analysis of the predicates, wherein the aggregate values are derived from the input data and characterize the input data;
process the aggregate values to calculate a plurality of key indicators based on the analysis of the predicates, wherein the key indicators are derived from the aggregate values and the key indicators represent a relationship between different aggregate values;
apply the inference rules to the aggregate values and key indicators to generate the advisories; and
to output at least some of the advisories on the operation of the business to the user.
2. The system of claim 1 , further comprising a data store for storing the plurality of aggregate values and the plurality of key indicators.
3. The system of claim 1 , wherein the plurality of inference rules comprise ‘if-then’ rules.
4. The system of claim 3 , wherein the processor determines rules for calculating the aggregate values and key indicators and stores the rules in the database.
5. The system of claim 3 , further comprising a rule editor, the rule editor providing a means of creating, editing, and saving the plurality of rules.
6. The system of claim 1 , wherein the advisory messages comprise natural language messages.
7. The system of claim 1 , wherein advisory messages are assigned a priority, and the processor is further programmed to prioritize the advisory messages.
8. The system of claim 1 , wherein the processor is further programmed to group the advisory messages into predetermined categories.
9. The system of claim 7 , wherein the priorities are assigned a priori to each advisory.
10. The system of claim 7 , wherein the priorities are determined responsive to a determination of the effect of a variation in an aggregate value or key indicator on a predefined metric.
11. A computer-implemented method of analyzing numerical data indicative of the operation of an enterprise and providing advisories to a user relevant the operation of the enterprise, comprising:
providing a plurality of inference rules regarding the operation of the enterprise;
analyzing the plurality of inference rules to determine key indicators relevant to the operation of the enterprise and to determine aggregate values needed for calculating the key indicators;
processing the numerical data to calculate the determined aggregate values, wherein the aggregate values are derived from the input data and the aggregate values characterize the input data;
processing, with a computer, the aggregate values to calculate the plurality of key indicators, wherein the key indicators are derived from the aggregate values and the key indicators represent relationships between different aggregate values;
comparing the key indicators with predefined thresholds or ranges; and
outputting advisories relevant to the operation of the enterprise responsive to the comparisons to the user.
12. The method of claim 11 , further comprising storing the plurality of aggregate values, the plurality of key indicators, and the predefined thresholds or ranges in a database.
13. The method of claim 11 , further comprising storing the plurality of inference rules in a rule database, wherein the comparisons are performed responsive to the inference rules.
14. The method of claim 13 , further comprising storing rules for calculating the aggregate values and key indicators in the rule database.
15. The method of claim 14 , further comprising deriving the rules for calculating the aggregate variables and key indicators from the inference rules.
16. The method of claim 13 , further comprising providing a means of creating, editing, and saving the plurality of inference rules.
17. The method of claim 11 , wherein the analytical results comprise advisory messages and the method further comprises selecting an advisory message responsive to the comparisons.
18. The method of claim 17 , wherein the advisory messages comprise natural language messages.
19. The method of claim 17 , wherein advisory messages have an associated priority, the further comprising prioritizing the advisory messages.
20. The method of claim 19 , wherein the advisory messages are grouped into predetermined categories.
21. The method of claim 20 , further comprising prioritizing the advisory messages within each category.
22. The method of claim 17 , wherein the priorities are assigned a priori to each advisory.
23. The method of claim 17 , further comprising determining the priorities responsive to a determination of the effect of a variation in an aggregate value or key indicator on a predefined metric.