Attribute-based classification and retrieval system

• US 5,201,047 A
• Filed: 12/21/1989
• Issued: 04/06/1993
• Est. Priority Date: 12/21/1989
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mukherjee, Sujan K., Maki, Reino A.
• Primary Examiner(s): Clark, David L.
• Assistant Examiner(s): MULLEN, THOMAS J
• Application Number:
  US07/452,227
• Time in Patent Office: 1,202 Days
• Field of Search: 364/513, 364/468, 364/900, 364/200
• US Class Current: 1/1
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

An attribute-based automated classification and retrieval system for group technology applications using a codeless classification system in which hierarchical classification structures are stored in a classification attribute file and in which searches can be performed at any level in the classification structure. The relationship between entities and classification attributes are kept in an item classification parameters file along with the parameter values associated with each entity-attribute pair. Queries on the data stored in the classification attribute file and item classification parameters file are stored alternately in a plurality of query results files as successive queries are used to narrow the scope of a search.

Claims

1. An attribute-based method of item classification and retrieval for group technology applications for implementation on a computer system wherein said computer system has an installed database system for storing database files including an item file on an external storage device and an interactive query system for retrieval of information stored in said database system in response to a query made by an operator at a workstation electrically connected to said computer system, said method comprising the steps of:

generating interactively at said workstation a plurality of multi-level classification attribute data structures with each data structure having at least one attribute identifier and, using said database system, storing said plurality of classification attribute data structures in a classification attributes file on said external storage device;

for each item stored in said item file, retrieving an associated item identifier and selecting interactively at said workstation a plurality of attribute identifiers from the plurality of attribute data structures stored in said classification attributes file pertaining to said item identifier to form a plurality of item identifier-attribute identifier pairs, and storing said plurality of item identifier-attribute identifier pairs in an item and classification attributes relationship file; and

entering a first query interactively by said operator at said workstation using said interactive query system to perform a search on at least one attribute identifier in said item and classification attributes file, retrieving each said item identifier that includes said at least one attribute identifier as part of the stored item identifier-attribute identifier pair, and, using said database system, storing each said item identifier retrieved by said query in a query results file.

2. The attribute-based method of item classification and retrieval of claim 1 including entering a second query interactively by said operator at said workstation using said interactive query system to perform a search on at least one attribute identifier in said classification attributes file, retrieving the multi-level classification attribute data structure that contains said at least one attribute identifier, and, using said database system, storing said retrieved multi-level classification attribute data structure in a query results file.

3. An attribute-based codeless item classification and retrieval system for group technology applications for implementation on a computer system wherein said computer system has an installed database system for storing database files including an item file on an external storage device and interactive query system for retrieval of information stored in said database system in response to a query made by an operator at a workstation electrically connected to said computer system, said codeless item classification and retrieval system comprising:

a first database table located on said external storage device for storing a plurality of multi-level classification attribute data structures, with each data structure having at least one attribute identifier;

a second database table located on said external storage device for storing a plurality of item identifier-attribute identifier pairs, said item identifier-attribute identifier paris being determined by said operator associating at least one attribute identifier from said first database table with each item identifier contained in said item file; and

a third database table located on said external storage device for storing the results of queries entered interactively by said operator at said workstation using said interactive query system and performed on said plurality of item identifier-attribute identifier pair stored in said second database table and on said plurality of multi-level classification attribute data structures stored in said first database table.

4. An attribute-based method of item classification and retrieval for group technology applications for implementation on a computer system wherein said computer system has an installed database system for storing database files including an item file on an external storage device and a query system for retrieval of information stored in said database system in response to a query made by an operator at a workstation electrically connected to said computer system, said method comprising the steps of:

generating interactively at said workstation a plurality of multi-level classification attribute data structures with each structure having at least one attribute identifier and, using said database system, storing said plurality of classification attribute data structures in a classification attributes file on said external storage device;

generating interactively at said workstation a plurality of modifiable templates with each template defining at least one parameter name associated with each attribute identifier stored in said classification attributes file and, using said database system, storing said plurality of modifiable templates in a classification parameter template file on said external storage device;

for each item stored in said item file, retrieving an associated item identifier and selecting interactively at said workstation a plurality of attribute identifiers from the plurality of attribute data structures stored in said classification attributes file pertaining to said item identifier to form a plurality of item identifier-attribute identifier pairs, and for each item identifier-attribute identifier pair determining a parameter value corresponding to each parameter name stored in said classification parameter template file that is associated with the attribute identifier, and storing said plurality of item identifier-attribute identifier paris along with said parameter value corresponding to each associated parameter name in an item classification parameters file; and

entering a first query interactively by said operator at said workstation using said interactive query system to perform a search on at least one attribute identifier in said item classification parameters file, retrieving each said item identifier-attribute identifier pair along with the parameter value corresponding to each associated parameter name that is associated with said at least one attribute identifier, and using said database system, storing each said item identifier-attribute identifier pair in one of a plurality of query results file.

5. The attribute-based method of item classification and retrieval of claim 4 including performing a series of interrelated queries on said item classification parameters file wherein a first query entered interactively by said operator at said workstation is made on a first attribute identifier to identify each item identifier that contains the first attribute in said plurality of item identifier-attribute identifier pairs with the results of the first query being stored in a first query results file using said database system, and a second query entered interactively by said operator at said workstation is made on a second attribute identifier wherein the scope of the search is limited to the item identifiers stored in said first query results file to identify those item identifiers in said first query results file that also have the second attribute with the results of the second query being stored in a second query results file using said database system.

6. The attribute-based method of item classification and retrieval of claim 5 wherein any subsequent searches are used to further limit the scope of the search to the item identifiers found in the immediately preceding search that have a specific parameter value corresponding to the associated parameter name.

7. The attribute-based method of item classification and retrieval of claim 5 including entering a second query interactively by said operator at said workstation using said interactive query system to perform a search on at least one attribute identifier in said classification attributes file, retrieving the multi-level classification attribute data structure that contains said at least one attribute identifier, and, using said database system, storing said retrieved multi-level classification attribute data structure in one of a plurality of query results files.

8. An attribute-based codeless item classification and retrieval system for group technology applications for implementation on a computer system wherein said computer system has an installed database system for storing database files including an item file on an external stores device and an interactive query system for retrieval of information stored in said database system in response to a query made by an operator at a workstation electrically connected to said computer system, said codeless item classification and retrieval system comprising:

a first database table located on said external storage device for storing a plurality of multi-level classification attribute data structures, with each structure having at least one attribute identifier;

a second database table located on said external stores device for storing a plurality of item identifier-attribute identifier pairs and a set of parameter values corresponding to each item identifier-attribute identifier pair, said item identifier-attribute identifier pairs being determined by said operator associating at least one attribute identifier from said first database table with each item identifier contained in said item file;

a third database table located on said external storage device for storing a plurality of parameter templates containing at least one parameter name associated with each classification attribute identifier stored in said first database table; and

a fourth database table located on said external storage device for storing the results of queries entered interactively by said operator at said workstation using said interactive query system and performed on said plurality of item identifier-attribute identifier pairs stored in said second database table and performed on said plurality of multi-level classification attribute data structures sores in said first database table.