Automatic taxonomy enrichment
AssigneeInternational Business Machines Corporation
InventorsSougata Mukherjea, Amit A. Nanavati, L. Venkata Subramaniam
Life of this patent
Estimating the term…
Abstract
Techniques for enriching a taxonomy using one or more additional taxonomies are provided. The techniques include receiving two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies, determining one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, importing one or more relevant portions from the one or more additional taxonomies into the destination taxonomy, and using the one or more imported taxonomy portions to enrich the destination taxonomy.
Claims10
A method for enriching a taxonomy using one or more additional taxonomies, wherein the method comprises:
- receiving two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- determining one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- importing the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- using the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic two-way enrichment, wherein said automatic two-way enrichment comprises accounting for each subsumed and conflicting match and synonym across the two taxonomies.
A computer program product comprising a non-transitory memory device including computer useable program code for enriching a taxonomy using one or more additional taxonomies, the computer program product including:
- computer useable program code for receiving two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- computer useable program code for determining one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- computer useable program code for importing the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- computer useable program code for using the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic two-way enrichment, wherein said automatic two-way enrichment comprises accounting for each subsumed and conflicting match and synonym across the two taxonomies.
A system for enriching a taxonomy using one or more additional taxonomies, comprising:
- a memory;
- and
- at least one processor coupled to the memory and operative to:
- receive two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- determine one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- import the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- use the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic two-way enrichment, wherein said automatic two-way enrichment comprises accounting for each subsumed and conflicting match and synonym across the two taxonomies.
A method for enriching a taxonomy using one or more additional taxonomies, wherein the method comprises:
- receiving two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- determining one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- importing the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- using the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic multi-way enrichment, wherein said automatic multi-way enrichment comprises accounting for each subsumed and conflicting match and synonym across three or more taxonomies, and ensuring that an order of one or more merges does not result in information loss.
A computer program product comprising a non-transitory memory device including computer useable program code for enriching a taxonomy using one or more additional taxonomies, the computer program product including:
- computer useable program code for receiving two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- computer useable program code for determining one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- computer useable program code for importing the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- computer useable program code for using the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic multi-way enrichment, wherein said automatic multi-way enrichment comprises accounting for each subsumed and conflicting match and synonym across three or more taxonomies, and ensuring that an order of one or more merges does not result in information loss.
A system for enriching a taxonomy using one or more additional taxonomies, comprising:
- a memory;
- and
- at least one processor coupled to the memory and operative to:
- receive two or more taxonomies, wherein the two or more taxonomies comprise a destination taxonomy and one or more additional taxonomies;
- determine one or more relevant portions of the two or more taxonomies by identifying one or more common terms between the two or more taxonomies, regardless of taxonomy intention, wherein said determining comprises:
- implementing a thesaurus to automatically recognize the one or more common terms and one or more relationships among the one or more common terms between the two or more taxonomies;
- and
- creating an order of the one or more common terms, wherein said order comprises a bottom-up order with respect to the destination taxonomy;
- import the one or more relevant portions from the one or more additional taxonomies into the destination taxonomy in accordance with said order of the one or more common terms while maintaining the unique perspective of the destination taxonomy via consideration of sub-tree enrichment and dependencies among matches in an order that is different from an order of tree enrichment associated with the destination taxonomy;
- and
- use the one or more imported taxonomy portions to perform asymmetric enrichment of the destination taxonomy by using one or more algorithms for automatic multi-way enrichment, wherein said automatic multi-way enrichment comprises accounting for each subsumed and conflicting match and synonym across three or more taxonomies, and ensuring that an order of one or more merges does not result in information loss.
Dependencies are inferred from the claim text.