Jonathan S Richards

US Patent:

6728700, Apr 27, 2004

Filed:

May 3, 1999

Appl. No.:

09/304461

Inventors:

Jonathan B. Richards - Lawrenceville GA
William H. Jones - Atlanta GA

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06F 1730

US Classification:

707 3, 707 4, 707102

Abstract:

A help interface capable of receiving user-defined queries in a natural language and selecting the most appropriate answer from a plurality of potential answers, includes facilities for parsing and storing the character string content of the query in a buffer. Lists of punctuation characters as well as single and multiple word character strings are successively compared with the content of the buffer to eliminate punctuation, superfluous words, location reference words, and to identify synonyms to base words meaningful in the selection of an appropriate answer. The identified base words are used as references into a predefined rule set which inherently matches the base word with a potential answer and assigns a value based on the relationship of the potential answer to the base word. The cumulative values of all potential answers are computed and ranked with the highest ranking value presented to the user as a response to the query.

US Patent:

59959215, Nov 30, 1999

Filed:

Apr 23, 1996

Appl. No.:

8/636663

Inventors:

Jonathan B. Richards - Lawrenceville GA
Joseph W. Guthridge - Atlanta GA
William H. Jones - Atlanta GA

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06F 1728
G06F 1730

US Classification:

704 9

Abstract:

A help interface capable of receiving user-defined queries in a natural language and selecting the most appropriate answer from a plurality of potential answers, includes facilities for parsing and storing the character string content of the query in a buffer. Lists of punctuation characters as well as single and multiple word character strings are successively compared with the content of the buffer to eliminate punctuation, superfluous words, location reference words, and to identify synonyms to base words meaningful in the selection of an appropriate answer. The identified base words are used as references into a predefined rule set which inherently matches the base word with a potential answer and assigns a value based on the relationship of the potential answer to the base word. The cumulative values of all potential answers are computed and ranked with the highest ranking value presented to the user as a response to the query.