John M Pratt

US Patent:

6697801, Feb 24, 2004

Filed:

Aug 31, 2000

Appl. No.:

09/652267

Inventors:

Michael A. Eldredge - Springville UT
Russell C. Johnson - Saratoga Springs UT
Ronald P. Millet - Orem UT
Bruce R. Tietjen - Highland UT
John P. Pratt - Pleasant Grove UT

Assignee:

Novell, Inc. - Provo UT

International Classification:

G06F 1730

US Classification:

707 6, 707100, 707102

Abstract:

Methods of hierarchically parsing and indexing text are provided. A source text is parsed to identify first level text entities and lower level text entities wherein the lower level text entities are subsets of their parent text entities. First level text entities are assigned a first level value corresponding to their locations within the source text, and lower level text entities are assigned lower level values, subordinate to their corresponding parent level value. Multiple levels of text entity parsing and numbering may take place. The association of linguistically derived variants of the text entities with their locations takes place.

US Patent:

6772141, Aug 3, 2004

Filed:

Dec 14, 1999

Appl. No.:

09/460059

Inventors:

John P. Pratt - Orem UT
Russell Clark Johnson - Saratoga Springs UT
Ronald P. Millett - Orem UT
Bruce R. Tietjen - Highland UT

Assignee:

Novell, Inc. - Provo UT

International Classification:

G06F 1730

US Classification:

707 3, 707 4, 707 7, 707102

Abstract:

A search decision table comprises a plurality of references to indexes, such as full text indexes, where each index corresponds to one or more informational resources. The search decision table has a plurality of references to ranges of text, wherein each range of text is bound by a lower text limit and an upper text limit. Preferably, the references to ranges of text are arranged in the matrix sequentially in alphanumeric order, and each reference to a range of text comprises the lower text limit. A plurality of cross-referencing data in the matrix correlate the references to the ranges of text and the references to the indexes, wherein each of the cross-referencing data correspond to a reference to a range of text and to a reference to an index. Preferably, the matrix includes attribute data and/or index data for each reference to range of text. Other embodiments of search decision tables are also disclosed, as well different uses for the search decision table.

US Patent:

7774353, Aug 10, 2010

Filed:

Aug 30, 2007

Appl. No.:

11/847869

Inventors:

Ronald P. Millett - Orem UT, US
Dillon K. Inouye - Provo UT, US
John C. Higgins - Orem UT, US
John P. Pratt - American Fork UT, US

Assignee:

Perfect Search Corporation - Provo UT

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707758, 707789

Abstract:

Searching a data store for parameter patterns specified in a query. A method includes receiving a query from a user including N parameter patterns. One or more alternatives are associated to one or more of the N parameter patterns. One or more templates are created. Each of the templates describes a number of microsearches. Each of the microsearches includes one or more of the N parameter patterns or the alternatives. Microsearches described by at least one of the one or more templates are enumerated. One or more sub-microsearches are performed by searching for parameter patterns and/or alternatives. Each sub-microsearch may have less than all terms needed for a full microsearch. Based on the results of the one or more sub-microsearches, one or more microsearches are eliminated from searching. The data store is searched using one or more of the remaining microsearches.

US Patent:

62161237, Apr 10, 2001

Filed:

Jun 24, 1998

Appl. No.:

9/103702

Inventors:

David O. Robertson - Provo UT
Ronald P. Millett - Orem UT
John P. Pratt - Orem UT
Bruce Tietjen - Highland UT

Assignee:

Novell, Inc. - Provo UT

International Classification:

G06F 1730

US Classification:

707 3

Abstract:

A method and system for generating and searching a full text index. The fill text index includes the use of word numbers and a minimum delta which minimizes the need to access document level information during the application of search operators. Word registers having coordinated document level and word level information, as well as relevance information are used in search operations. Word numbers are clustered together during sub-operations in preparation for the next operation in a search query. The fill text index according to the present invention is extremely efficient and greatly reduces table accesses and/or disk I/Os.