Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2002-09-30
2004-06-08
Mizrahi, Diane D. (Department: 2175)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000
Reexamination Certificate
active
06748376
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to software for database interrogation and manipulation and, more particularly, to a method and system for retrieving database records using predefined classifications, and further coupled with search refinement options based on parametrics and classification.
BACKGROUND OF THE INVENTION
Searchable electronic catalogs are commonly used in support of various electronic commerce and purchasing functions. These catalogs must have a user interface for selectively retrieving data records. Software engineers desire to make the user interfaces as simple as possible to operate, because complexity of the user interface can be a detriment to sales from the catalog. Simplicity becomes particularly important when the catalog is intended to be accessed by users with varying levels of skill or training.
User interfaces that are simple to operate should have the capability to handle almost any type of user input. In the case of an electronic catalog, if the user knows the exact part number of the desired product and enters the part number correctly into the user interface, then the database search engine will quickly identify the desired record from the database based on an exact match with the search string. In a more general case, the user may have only partial information about the desired product, or may incorrectly type the search string.
Danish et al. in U.S. Pat. No. 5,715,444 disclose a process for identifying a single item from a family of items in a database. A feature screen and search process present the user with a guided nonhierarchical parametric search to identify matching items based upon user specified criteria and priorities. Also disclosed are a method and system appropriate in an Internet environment.
Cochran et al. in U.S. Pat. Nos. 4,879,648 and 5,206,949 disclose a method of variably displaying search terms in which two control inputs are used to select a plurality of terms for a plurality of categories. A term in a visible position on the screen becomes a search term or a qualifier for other records in the database. The search results are dynamically formed on the basis of selected search terms. The search results can also be grouped in fixed or static lists.
Blutinger et al. in U.S. Pat. No. 5,231,566 disclose a method and apparatus for producing a catalog. The catalog has the characteristic that all like items in the database have the same catalog item number, whereas different items have different catalog item numbers.
Geier et al. in U.S. Pat. No. 4,984,155 disclose a system for enabling a customer to operate a data terminal for placing an order for goods or services. The customer is permitted to enter an invalid catalog number that is used as a key to display a series of products having catalog numbers similar to the catalog number entered by the customer.
Prior catalog search algorithms typically employ one of two search strategies. The first strategy is a keyword search for selecting database records based on matching text strings. The second strategy is a hierarchical search for selecting database records based on lists of classifications from which to narrow and select the database records. Each of the two search strategies has disadvantages that can make it difficult for users to find their desired database records.
The keyword search strategy has the disadvantage that users must be familiar with the appropriate key word terms that are likely to yield the desired data records. In addition, it is not always possible to quickly collect groups of logically related data records. If a close match is found, but it is not the desired exact match, it is not always possible to utilize the information in the close match to quickly identify all similar data records. A keyword search engine does not typically have a “more-like-this” function that operates on close matches to identify similar items within the database.
The hierarchical search strategy can take advantage of a logical grouping of data records. This search strategy is best suited for finding data that break down logically into successively greater levels of detail. This search strategy is most effective when the data have been carefully edited and structured within a database. Finding a single relevant record can quickly lead to all other relevant records, as long as the grouping logic relates to the way in which the data are used.
Thus, a “more-like-this” function can quickly identify all similarly classified records in the database.
The disadvantage of the hierarchical search strategy is that users may not always anticipate the proper classification of certain records, and may search the wrong categories for their desired database record. The user is tied to the logical structure of the data, and must learn to navigate the predefined structure of the database in order to locate particular data records.
It would be desirable to allow free-form text searching, with no prerequisites for format or content. Thus, it would be desirable to have a system capable of identifying the database records most likely to be the desired choice of the user, even when the user inputs a search string having misspelled terms, word fragments, or other characteristics of the item being sought. It would further be desirable to take advantage of natural relationships and logical groupings within the data records to compile lists of similarly classified data records. The prior art has not disclosed a system that provides a simple and convenient user interface coupled with a search engine that has the architecture and advantages of the present invention.
SUMMARY OF THE INVENTION
The present invention provides a simple user interface that combines the ease of keyword searching with the advantages of search refinement through classification or parametric selections. The text searching is improved through the use of sequential search algorithms that are designed to maximize the chances of identifying the desired data records. The search refinement is presented as a simple selection from a list of classifications that is dynamically compiled based on the results of the keyword search. The output to the user reports on both generic and category-specific product characteristics.
According to the present invention, a method of selecting data records in a catalog database comprises the following steps: inputting search terms to a user interface; testing the search terms against the catalog using a sequence of search algorithms, wherein each search algorithm is applied against the database to identify matching catalog records comprising a set; terminating the sequence of search algorithms when at least one record becomes a member of the set; compiling a unique list of classifications from each member of the set, said list comprising at least a first classification; displaying at least a portion of the set along with the list of classifications; creating a subset of the set by selecting members having the first classification; and displaying the subset.
The invention comprises a database along with a search engine. The database typically consists of category descriptions, manufacturer's name, manufacturer part number, short text description, and parametrically composed descriptions. Product categories and characteristics are represented directly by tables and columns respectively.
The search engine executes a series of text string matching algorithms, in sequence, until at least one match is identified from the catalog. The sequence comprises proximity searching, string matching, stemming, fuzz logic, and soundex matching. For example, if an exact match is found, the search halts when all exact matches have been identified, and there is no further recourse to other search algorithms. If no exact match is found, then the search terms are manipulated to identify strings with similar roots. If, again, no match is found, the search terms are tested further according to other algorithms, such as fuzzy logic and soundex, until a match is found or the search engine reaches its logical termi
Beall Christopher Wade
Neal Michael Renn
Wilmsen James Michael
Holland & Hart LLP
Kulish, Esq. Christopher J.
Mizrahi Diane D.
Requisite Technology Inc.
LandOfFree
Method and system for database manipulation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for database manipulation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for database manipulation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3299576