Rule based expert system for consumer preference
First Claim
1. A knowledge-based system for providing a recommendation tailored to a consumer, the system comprising:
- a knowledge base containing historical data;
rule extraction means for extracting a ruleset from the knowledge base;
a rules database for holding the ruleset;
codifying means for codifying the requirements of the consumer; and
recommendation means for applying the ruleset to the codified consumer requirements and generating the recommendation accordingly;
explanation means for explaining to the consumer at least one reason for the recommendation; and
means for using XML (Extensible Markup Language) to define consumers and possible/actual recommendations.
7 Assignments
0 Petitions
Accused Products
Abstract
An expert system and a method of providing automated advice are described. The system is regularly updated by advice (or diagnosis, recommendation etc.) given by practitioners in the relevant field. The combination of the underlying facts and the consequent (human) advice is used to update a ruleset that is then used to provide automated advice. An example of financial advice is given. A database contains the details for the customers of a financial institution together with advice and recommendations given previously by the institution'"'"'s human advisors. This database is used to derive a ruleset which is then applied to a subsequent customer'"'"'s details in an automated manner, possibly at the user'"'"'s own PC via the World Wide Web. Frequent updating using data from the human advisors'"'"' work means that the database (and hence the ruleset) are kept up to date. Consistent advice can thus be provided with minimum human interaction.
111 Citations
39 Claims
-
1. A knowledge-based system for providing a recommendation tailored to a consumer, the system comprising:
-
a knowledge base containing historical data;
rule extraction means for extracting a ruleset from the knowledge base;
a rules database for holding the ruleset;
codifying means for codifying the requirements of the consumer; and
recommendation means for applying the ruleset to the codified consumer requirements and generating the recommendation accordingly;
explanation means for explaining to the consumer at least one reason for the recommendation; and
means for using XML (Extensible Markup Language) to define consumers and possible/actual recommendations. - 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. A method of operating a knowledge-based system to provide a recommendation tailored to a consumer, the method comprising:
-
extracting rules from a knowledge base; and
applying the extracted rules to codified consumer requirements to generate the recommendation accordingly and explaining to the consumer at least one reason for the recommendation, wherein XML (Extensible Markup Language) is used to define consumers and possible/actual recommendations. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39)
explaining to the consumer the rules on which the recommendation was based.
-
-
27. A method according to claim 24, further comprising:
expressing the explanation substantially in a natural language.
-
28. A method according to claim 24, wherein Lozowski'"'"'s algorithm is applied to extract rules from the knowledge base.
-
29. A method according to claim 24, further comprising suggesting a plurality of alternative recommendations.
-
30. A method according to claim 29, further comprising ranking recommendations by their suitability.
-
31. A method according to claim 24, further comprising updating the knowledge base with changing details of recommendations that can be made.
-
32. A method according to claim 24, wherein the codified consumer requirements include the perceived needs of the consumer and the circumstances of the consumer.
-
33. A method according to claim 24, further comprising storing a consumer'"'"'s details for later recall.
-
34. A method according to claim 33, wherein the consumer'"'"'s details are stored and later recalled when providing a future recommendation.
-
35. A method according to claim 33, wherein the consumer'"'"'s details are stored and later recalled when completing a recommendation, input of consumer requirements being suspended in the meantime.
-
36. A method according to claim 24, wherein the consumer obtains advice from a remote human adviser.
-
37. A method according to claim 36, wherein the advice is obtained over a video conference link.
-
38. A method according to claim 37, wherein the consumer enters consumer requirements via a consumer terminal being a home PC or a kiosk, booth, ATM or other terminal in a financial advice establishment.
-
39. A method according to claim 38, wherein the consumer enters consumer requirements in response to a sequence of questions, and wherein later questions in the sequence are adapted in accordance with answers given to earlier questions in the sequence.
Specification