Ontological concept expansion

Patent metadata

Field	Value
Publication number	US-9230215-B2
Application number	US-201313869364-A
Country	US
Kind code	B2
Filing date	Apr 24, 2013
Priority date	Apr 24, 2013
Publication date	Jan 5, 2016
Grant date	Jan 5, 2016

Abstract

In one embodiment, prior to similarity measure computation, concept expansion based on disjunctive normal form (DNF) decomposition and non-conventional reasoning is applied to an original ontology to generate an expanded ontology having the original concepts plus one or more pseudo concepts. As a result, the similarity measure computation can produce results that more accurately reflect a human point of view than convention techniques.

First claim

What is claimed is:

1. A machine-implemented method comprising:
(a) the machine receiving an original ontology containing a plurality of original concepts; and
(b) the machine applying a concept expansion mechanism to the original ontology to generate an expanded ontology containing the original concepts and one or more pseudo concepts, wherein step (b) comprises:
(b1) generating an in-memory representation for the original ontology;
(b2) using one or more libraries of reasoning rules to generate the one or more pseudo concepts; and
(b3) incorporating the one or more pseudo concepts into the in-memory representation.

2. The method of claim 1 , further comprising:
(c) the machine applying a similarity measure computation to the expanded ontology to characterize similarity between at least two concepts in the expanded ontology.

3. The method of claim 1 , wherein step (b) further comprises retrieving a list of the reasoning rules from a configuration file.

4. The method of claim 3 , further comprising modifying the configuration file to enable retrieval of a different list of the reasoning rules.

5. The method of claim 1 , wherein step (b1) comprises using a semantic web reasoner to generate the in-memory representation for the original ontology.

6. The method of claim 1 , wherein, for at least one library of reasoning rules:
step (b2) comprises applying a current in-memory representation corresponding to an ontology containing the original concepts and one or more previously added pseudo concepts to the library to generate one or more new pseudo concepts; and
step (b3) comprises adding the one or more new pseudo concepts to the current in-memory representation to generate an updated in-memory representation containing the original concepts, the one or more previously added pseudo concepts, and the one or more new pseudo concepts.

7. The method of claim 6 , wherein step (b3) comprises using a semantic web reasoner to sequentially add each of the one or more new pseudo concepts to the current in-memory representation.

8. A machine-implemented method comprising:
(a) the machine receiving an original ontology containing a plurality of original concepts; and
(b) the machine applying a concept expansion mechanism to the original ontology to generate an expanded ontology containing the original concepts and one or more pseudo concepts, wherein step (b) comprises applying at least one non-conventional reasoning process to a logical expression for a concept to generate at least one new pseudo concept.

9. The method of claim 8 , wherein step (b) comprises using a semantic web reasoner to generate the logical expression for the at least one new pseudo concept.

10. The method of claim 8 , wherein the at least one non-conventional reasoning process (i) rewrites the logical expression for the concept in negative normal form and then in disjunctive normal form and (ii) generates a first new pseudo concept for at least one disjunct in the logical expression written in the disjunctive normal form.

11. The method of claim 10 , wherein the at least one non-conventional reasoning process (iii) generates a second new pseudo concept for at least one conjunct used to generate the first new pseudo concept.

12. The method of claim 8 , wherein the at least one non-conventional reasoning process (i) uses least common subsumer reasoning to determine a least common subsumer of at least one pair-wise disjoint concept and (ii) generates a new pseudo concept for the least common subsumer.

13. The method of claim 8 , wherein the at least one non-conventional reasoning process uses super-properties of the logical expression in order to deduce a new pseudo concept as a less-strict concept.

14. The method of claim 8 , wherein the at least one non-conventional reasoning process applies one or more distributive rules for one or more existential restrictions in the logical expression to generate a new pseudo concept.

15. The method of claim 8 , wherein the at least one non-conventional reasoning process applies one or more distributive rules for one or more universal restrictions in the logical expression to generate a new pseudo concept.

16. The method of claim 8 , wherein the at least one non-conventional reasoning process applies one or more distributive rules for one or more cardinality restrictions in the logical expression to generate a new pseudo concept.

17. The method of claim 8 , wherein the at least one non-conventional reasoning process applies a closed world assumption to the logical expression to generate a new pseudo concept.

18. The method of claim 17 , wherein, if the concept has (i) a non-null intersection with another concept and (ii) all its members into this intersection, then the concept's definition is modified to assert that the concept is equivalent to all its members.

19. The method of claim 8 , wherein the at least one non-conventional reasoning process derives a universal clause from an existential clause in the logical expression to generate a new pseudo concept.

20. The method of claim 8 , further comprising:
(c) the machine applying a similarity measure computation to the expanded ontology to characterize similarity between at least two concepts in the expanded ontology.

21. A machine comprising:
an input node configured to receive an original ontology containing a plurality of original concepts; and
a concept expansion mechanism configured to apply concept expansion the original ontology to generate an expanded ontology containing the original concepts and one or more pseudo concepts, wherein the concept expansion mechanism applies at least one non-conventional reasoning process to a logical expression for a concept to generate at least one new pseudo concept.

Assignees

• Alcatel Lucent
• Alcatel Lucent

Inventors

• Christophe Benoit
• Boussard Mathieu

Classifications

• H04L29/08756

  Electricity · mapped topic

• G06N99/005

  Physics · mapped topic

• G06F17/30707

  Physics · mapped topic

• G06F17/3074

  Physics · mapped topic

• G06F17/30914

  Physics · mapped topic