Adaptive travel purchasing optimization system

• US 6,477,520 B1
• Filed: 02/22/1999
• Issued: 11/05/2002
• Est. Priority Date: 02/22/1999
• Status: Expired due to Term

• Current Assignee: Datalex Communications USA, Inc.
• Original Assignee: Yatra Corporation (Datalex Plc)
• Inventors: Malaviya, Paritosh, Saklani, Praful, Saklani, Mukul, Malaviya, Ashutosh
• Primary Examiner(s): STARKS, WILBERT L
• Application Number:
  US09/255,178
• Time in Patent Office: 1,352 Days
• Field of Search: 704/8, 706/2, 706/13, 706/25, 706/47, 706/58
• US Class Current: 706/47
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• 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

Abstract

A travel purchasing optimization system consists of a cognizer in communication with a knowledge base and a rule base. A knowledge acquisition and rule manager (KARM) dynamically creates, alters, and adaptively fine tunes rules for the rule base from existing business knowledge using a rule modeling language. A travel request is made through an external booking system and sent to the cognizer via an integration framework and the knowledge base. The integration framework pools data from a series of external systems or databases and the cognizer compares the rules from the rule base with the pooled data from the integration framework. The cognizer generates the total cost of travel (TCOT) and sends the TCOT back to the external booking system via the integration framework.

Claims

1. A computer-implemented travel purchasing optimization system (TPOS) for providing choices for business travel in response to queries, wherein the choices are based on a computation of a total cost of travel, the system comprises:

a cognizer in communication with a knowledge base, a rule base and a knowledge acquisition and rule manager (KARM), wherein said knowledge acquisition and rule manager creates rules, including fuzzy logic rules, for said rule base using a rule modeling language, and further wherein, an external booking system transmits a travel request to an integration framework in communication with said knowledge base, and said integration framework collects data from a plurality of system connectors, and further wherein, said cognizer processes said rules, including said fuzzy logic rules, from said rule base with the data from said knowledge base so as to determine a total cost of travel in response to the travel request.

2. The computer-implemented travel purchasing optimization system of claim 1 wherein the rule modeling language comprises an adaptive rule modeling language.

3. The computer-implemented travel purchasing optimization system of claim 1 wherein said knowledge acquisition and rule manager further comprises:

(a) a rule manager for creating and managing said rules based on a plurality of preferences;

(b) a rule writer in communication with said rule manager and said rule base for creating and managing said rules based on said rule modeling language;

(c) a rule tester in communication with said rule manager, said rule writer, and said rule base for testing said rules; and

(d) a rule analyzer in communication with said rule manager, said rule tester, and said knowledge base for analyzing said rules.

4. The travel purchasing optimization system of claim 1 wherein said rule modeling language further comprises:

(a) a terminal vocabulary;

(b) a non-terminal vocabulary;

(c) a plurality of production rules;

(d) a starting symbol;

(e) a meta terminal vocabulary;

(f) a meta non-terminal vocabulary;

(g) a plurality of meta rules; and

(h) a meta starting symbol wherein said language components (a)-(h) combine to form adaptive rules.

5. The computer-implemented travel purchasing optimization system of claim 1 wherein said rule base comprises:

(a) at least one rule block comprising a plurality of said rules for processing data; and

(b) a plurality of clusters, each of the plurality of clusters comprising at least one rule block for processing data.

6. The computer-implemented travel purchasing optimization system of claim 1 wherein said system connectors further comprise:

(a) a corporate preference editor in communication with said integration framework for providing travel preferences of the corporation;

(b) an employee preference editor in communication with said integration framework for providing travel preferences of the employee;

(c) a special pricing database for providing corporate contract information; and

(d) a plurality of external databases including human resources profiles, computer reservation systems, and custom applications.

7. The travel purchasing optimization system of claim 1 wherein said cognizer further comprises:

(a) a reader in communication with said rule base for reading and compiling said rule modules and said core modules;

(b) an object populator in communication with said knowledge base for creating a plurality of global objects;

(c) a cognifier/decognifier (CODEC) in communication with said object populator and said YDL reader for relating said rule modules and said global objects;

(d) a linguistic function library in communication with said CODEC for supplying fuzzy logic functions to said CODEC;

(e) a confidence analyzer in communication with said linguistic function library for testing and monitoring said linguistic function library;

(f) a process manager in communication with said CODEC and said confidence analyzer for managing said cognizer and initializing said CODEC and said global objects;

(g) a rules firing engine in communication with said process manager for creating a plurality of weight sets for said rule clusters; and

(h) a total cost computation module in communication with said rules firing engine and said process manager for calculating the total cost of travel.

8. The travel purchasing optimization system of claim 1 wherein said travel purchasing optimization system further comprises a data mart in communication with said knowledge acquisition and rule manager for creating and archiving travel purchasing optimization system data.

9. A method of travel purchase optimization comprising the steps of:

(a) creating a plurality of rules, including fuzzy rules, using a knowledge acquisition and rule manager (KARM);

(b) receiving data related to a travel request;

(c) calculating a total cost of travel by applying the plurality of rules to the received data; and

(d) interfacing with an external booking system using an application program interface in communication with said integration framework.

10. The method of claim 9 further comprising the steps of:

a) receiving said received data through a plurality of system connectors;

b) compiling the received data using an integration framework in communication with said system connectors; and

c) storing said received data in a knowledge base in communication with said integration framework.

11. The method of claim of claim 9 wherein the step of creating rules using said knowledge acquisition and rule manager further comprises the steps of:

(a) creating said rules based on a plurality of preferences using a rule manager;

(b) writing said rules in a rule modeling language using a rule writer in communication with said rule base;

(c) testing said rules using a rule tester in communication with said rule manager, said rule writer, and said rule base; and

(d) analyzing said rules using a rule analyzer in communication with said rule manager, said rule tester, and said knowledge base.

12. The computer-implemented travel purchasing optimization system of claim 11 wherein the rule modeling language comprises an adaptive rule modeling language.

13. The travel purchasing optimization system of claim 11 wherein writing said rules in said rule modeling language using a rule manager further comprises the steps of:

(a) coding a terminal vocabulary;

(b) coding a nonterminal vocabulary;

(c) coding a plurality of production rules;

(d) coding a starting symbol;

(e) coding meta terminal vocabulary;

(f) coding a meta non-terminal vocabulary;

(g) coding a plurality of meta rules; and

(h) coding a meta starting symbol.

14. The travel purchasing optimization system of claim 9 further comprising the steps of:

a) storing said rules in said knowledge base;

b) coding at least one rule block, each comprising a plurality of said rules; and

c) coding a plurality of clusters each comprising at least one of said rule blocks.

15. The travel purchasing optimization system of claim 10 wherein using said system connectors comprises the steps of:

(a) connecting to a special pricing database to provide corporate contract information; and

(b) connecting to a plurality of external databases to provide human resource profiles, flight and fare information, and financial data.

16. The travel purchasing optimization system of claim 9 wherein the step of creating a plurality of rules further comprises the steps of:

a) connecting to a corporate preference editor to provide travel preferences of the corporation;

b) connecting to an employee preference editor to provide travel preferences of the employee; and

c) translating said corporation and employee preferences into a rule modeling language.

17. The travel purchasing optimization system of claim 9 wherein calculating the total cost of travel further comprises the steps of:

(a) using said cognizer for reading said plurality of rule blocks and said clusters using a descriptive language reader in communication with said rule base;

(b) compiling said rule modules and! said core modules using a descriptive language reader in communication with said rule base;

(c) creating a plurality of global objects using an object populator in communication with said knowledge base;

(d) relating said rule modules and said global objects using a cognifier/decognifier (CODEC) in communication with said object populator and said descriptive language reader;

(e) supplying fuzzy logic functions to said CODEC using a linguistic function library to in communication with said CODEC;

(f) monitoring said linguistic function library using a confidence analyzer in communication with said linguistic function library;

(g) managing said cognizer using a process manager;

(h) initializing said CODEC and said global objects using said process manager in communication with said CODEC and said confidence analyzer;

(i) creating a plurality of weight sets for said rule modules using a rules firing engine in communication with said process manager; and

(j) calculating the total cost of travel using a total cost computation module in communication with said rules firing engine and said process manager.

18. The method of travel purchase optimization of claim 17 further comprising the step of archiving information in a data mart in communication with said knowledge acquisition and rule manager.