Establishing semantic equivalence between concepts

Patent metadata

Field	Value
Publication number	US-10535003-B2
Application number	US-201415022870-A
Country	US
Kind code	B2
Filing date	Sep 22, 2014
Priority date	Sep 20, 2013
Publication date	Jan 14, 2020
Grant date	Jan 14, 2020

Abstract

A method for establishing semantic equivalence between a plurality of concepts including: providing an Orthogonal Semantic Equivalence Map in which first, second, and third extensional concept models are related; selecting or de-selecting a concept in the first concept model; selecting or deselecting a (relation, concept) pair representing an intensional relation from a concept in the first concept model to a concept in the second concept model over a concept in the third concept model; determining a subset of intensional relations from the selected concepts in the first concept model to concepts in the second concept model; determining a set of concepts from the first concept model that are related to concepts in the second concept model over the selected (relation, concept) pairs; and determining the narrowest common extension of the set of concepts from the first, second, or third concept models that are related over the selected intensional relations.

First claim

What is claimed is:

1. A method for establishing semantic equivalence between a plurality of concepts, comprising the steps of:
a. providing an Orthogonal Semantic Equivalence Map in which first, second, and third extensional concept models are related to one another such that the second concept model is orthogonal to the first concept model and the third extensional concept model is distinct from the first and second concept models, wherein each concept from the first concept model has an intensional relation to one concept from the second concept model over one concept in the third concept model as a (relation, concept) pair, wherein each concept represented in the first concept model is selectable or de-selectable, and wherein each intensional relation between concepts in the first and second concept model is selectable or de-selectable;
b. at least one of selecting or de-selecting a concept in the first concept model;
c. at least one of selecting or de-selecting a (relation, concept) pair representing an intensional relation from a concept in the first concept model to a concept in the second concept model over a concept in the third concept model;
d. based on the at least one of selecting or de-selecting a concept in the first concept model, determining a subset of intensional relations from the selected concepts in the first concept model to concepts in the second concept model;
e. based on the at least one of selecting or de-selecting a (relation, concept) pair representing and intensional relation over a concept in the third concept model, determining a set of concepts from the first concept model that are related to concepts in the second concept model over the selected (relation, concept) pairs,
f. based on the set of selected (relation, concept) pairs, determining a set of deselected (relation, concept) pairs; and
g. determining at least one of the narrowest common extension of the set of concepts from the first, second, or third concept models that are related over the selected intensional relations, wherein the narrowest common extension of the selected concepts from the first concept model is designated as being semantically equivalent to the set of selected (relation, concept) pairs relating each selected concept from the first concept model to a concept in the second concept model.

2. The method of claim 1 , further comprising simplifying the set of selected (relation, concept) pairs comprising substituting a plurality of (relation, concept) pairs with a single (relation, concept) pair comprising a relation over the narrowest common extension of the subset of concepts from the third concept model represented in the subset of (relation, concept) pairs and the narrowest common extension of the subset of concepts from the second concept model represented in the subset of (relation, concept) pairs.

3. The method of claim 1 , further comprising processing the Orthogonal Semantic Equivalence Map to produce a plurality of assertions comprising at least one of the set of selected (relation, concept) pairs and the set of de-selected (relation, concept) pairs.

4. The method of claim 3 , wherein the description of one characteristic of an entity comprises a set of assertions of intensional properties of the entity, wherein each assertion includes one of:
a. a concept in the first concept model;
b. a name, term, label, phrase, or identifier for a concept in the first concept model;
c. a tuple having a format (entity, (relation, concept)), wherein the concept is in the second concept model, and wherein the relation is a relation over a concept in the third concept model;
d. a tuple having a format (entity, (relation, function (parameters))), wherein the function maps the parameters to a concept in the second concept model, and wherein the relation is a relation over a concept in the third concept model;
e. a tuple having a format (entity, (relation, text)), wherein the text is a name, term, label, phrase, or identifier for a concept in the second concept model, and wherein the relation is a relation over a concept in the third concept model; or
f. a tuple having a format (entity, (first text, second text)), wherein the first text is a name, term, label, phrase, or identifier for a concept in the third concept model, and wherein the second text is a name, term, label, phrase, or identifier for a concept in the second concept model;
the method further comprising
g. constructing a concept filter having a second Orthogonal Semantic Equivalence Map,
h. using the concept filter, applying the set of assertions of intensional properties to the second Orthogonal Semantic Equivalence Map to produce an intersection of the selected concepts from the first concept models for the first Orthogonal Semantic Equivalence Map and the second Orthogonal Semantic Equivalence Map,
i. using the concept filter, determining the narrowest common extension of the intersection, and
j. if the narrowest common extension is not a top level concept, designating the entity as matching the concept filter over the first Orthogonal Semantic Equivalence Map.

5. The method of claim 4 wherein the concept filter is assigned a unique identifier, stored in a structured format in a storage medium, and is retrievable using the unique identifier.

6. The method of claim 5 , further comprising a corpus comprising a plurality of electronic resources each comprising textual content, wherein, for each electronic resource in the corpus, the method further comprises
a) using one or more Information Extraction systems to recognize within the textual content at least one name, term, label, or identifier of an entity and at least one name, term, label, or identifier of concepts from the first, second, and third concept set corresponding to the entity and to produce a set of assertions for the entity, and
b) resolving the concepts represented by the identified names, terms, labels, or identifiers to semantically equivalent concepts in a target concept set selected from the first, second, and third concept sets.

7. The method of claim 6 , further comprising modifying the textual content by substituting the at least one name, term, label, or identifier identified in a) with a name, term, label, or identifier mapped to semantically equivalent concepts identified in b).

8. The method of claim 7 , further comprising modifying the textual content by embedding a tag, attribute, link, or metadata comprising a name, term, label, or identifier mapped to semantically equivalent concepts identified in b).

9. The method of claim 8 , further comprising producing a list comprising all semantically equivalent concepts identified in the textual content.

10. The method of claim 9 , further comprising
determining a narrowest common extension of an intersection of a list of semantically equivalent concepts identified in textual content of a first electronic document and a list of semantically equivalent concepts identified in textual content of a second electronic document, and
using the resulting concept, placing the first and second electronic documents in a group identified by the resulting concept.

11. The method of claim 10 , further comprising storing a persistent identifier of one or more concept filters in an index with the electronic documents in which the semantically equivalent concepts were identified.

12. The method of claim 11 , further comprising creating a second concept filter by applying the first set of assertions comprising terms, concepts, or tuples of concepts and relations to select or de-select semantically equivalent concepts in a second Orthogonal Semantic Equivalence Map, and storing the second concept fil

Assignees

• Namesforlife Llc
• Univ Michigan State

Inventors

• Parker Jr Charles T
• Garrity George M
• Krdzavac Nenad

Classifications

• G06F16/2228

  Indexing structures · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title

• G06F16/23

  Updating · CPC title

• G06N5/02 Primary

  Knowledge representation; Symbolic representation · CPC title