Relativistic concept measuring system for data clustering

1. A method for creating a relativistic metric for measuring distances between sub-structures in an ontology implemented using a computer having a microprocessor, comprising:

setting, using the microprocessor, a fixed distance, between a top of a first order structure and each sub-structure in the order structure, to a constant value independent of the depth of the order structure, thereby generating relativity between sub-structures, the sub-structures corresponding to concepts, relationships or attributes of an entity;

creating, using the microprocessor, a model of the ontology based on a hierarchy of the sub-structures in the order structure using the fixed distance and a plurality of factors forming the basis of a relativistic conceptual distance metric;

obtaining a data set;

generating, using the microprocessor, a second order structure for the data set;

generating, using the microprocessor, a semantic distance field for clustering by applying the second order structure and the data set to the model of the ontology created in the creating step, the creating step further creating the model of the ontology by mapping concepts and features to create a vector-valued partial order structure based on the plurality of factors forming the basis of the relativistic conceptual distance metric, computing relative conceptual distances using the vector-valued partial order structure, and defining a Semantic Distance Field model based on the relative conceptual distances;

performing data clustering based on the generated semantic distance field in an N dimensional space; and

inducing, using the microprocessor, the ontology from the data clustering performed by the performing step.

2. The method according to claim 1 , wherein the step of mapping concepts and features to create the vector-values partial order structure includes the step of

mapping concepts and features to Rvachev-functions to create the vector-values partial order structure.

3. The method according to claim 2 , wherein the step of mapping concepts and features to Rvachev-functions to create the vector-values partial order structure, further includes:

generating, for each object of a plurality of objects, equations representing boundaries of attributes for each respective object;

converting, for each object of a plurality of objects, the equations into greater than or equal to zero type inequalities;

generating, for each object of a plurality of objects, a logical expression combining regions of space defined by the inequalities into a semantic entity;

substituting, for each object of a plurality of objects, the logical expression with a corresponding rvachev-function such that the resulting rvachev-function is equal to 0 on a boundary of the semantic entity, greater then 0 inside a region of the semantic entity, and less then 0 outside the region of the semantic entity; and

generating a composite rvachev-function representing logical statements corresponding to the plurality of objects using the respective rvachev-functions of the objects.

4. The method according to claim 1 , wherein the step of computing relative conceptual distances using the vector-valued partial order structure further includes using a potential function or a virtual gravity function.

5. The method according to claim 1 , wherein the plurality of factors forming the basis of the relativistic conceptual distance metric include one of a content factor, a relevancy factor, a density Factor, a depth factor, relationship type factor and a directed order factor.

6. An apparatus for creating a relativistic metric for measuring distances between sub-structures in an ontology, comprising:

a computer having a microprocessor implementing:

a setting unit configured to set a fixed distance, between a top of a first order structure and each sub-structure in the order structure, to a constant value independent of the depth of the order structure, thereby generating relativity between sub-structures, the sub-structures corresponding to concepts, relationships or attributes of an entity,

a creating unit configured to create a model of the ontology based on a hierarchy of the sub-structures in the order structure using the fixed distance and a plurality of factors forming the basis of a relativistic conceptual distance metric,

an obtaining unit configured to obtain a data set,

a first generating unit configured to generate a second order structure for the data set,

a second generating unit configured to generate a semantic distance field for clustering by applying the second order structure and the data set to the model of the ontology created in the creating step, wherein the creating unit is further configured to create the model of the ontology by mapping concepts and features to create a vector-valued partial order structure based on the plurality of factors forming the basis of the relativistic conceptual distance metric, compute relative conceptual distances using the vector-valued partial order structure, and define a Semantic Distance Field model based on the relative conceptual distances;

a data clustering unit configured to perform data clustering based on the generated semantic distance field in an N dimensional space; and

an ontology unit configured to induce the ontology from the data clustering performed by the data clustering unit.

7. The apparatus according to claim 6 , wherein the creating unit is further configured to map concepts and features to create the vector-values partial order structure by mapping concepts and features to Rvachev-functions to create the vector-values partial order structure.

8. The apparatus according to claim 7 , wherein the creating unit is further configured to map concepts and features to Rvachev-functions to create the vector-values partial order structure by being configured to

generate, for each object of a plurality of objects, equations representing boundaries of attributes for each respective object,

convert, for each object of a plurality of objects, the equations into greater than or equal to zero type inequalities,

generate, for each object of a plurality of objects, a logical expression combining regions of space defined by the inequalities into a semantic entity,

substitute, for each object of a plurality of objects, the logical expression with a corresponding rvachev-function such that the resulting rvachev-function is equal to 0 on a boundary of the semantic entity, greater then 0 inside a region of the semantic entity, and less then 0 outside the region of the semantic entity, and

generate a composite rvachev-function representing logical statements corresponding to the plurality of objects using the respective rvachev-functions of the objects.

9. The method according to claim 6 , wherein the creating unit is further configured to compute relative conceptual distances using the vector-valued partial order structure further includes using a potential function or a virtual gravity function.

10. The apparatus according to claim 6 , wherein the plurality of factors forming the basis of the relativistic conceptual distance metric include one of a content factor, a relevancy factor, a density Factor, a depth factor, relationship type factor and a directed order factor.

11. A non-transitory computer readable medium having stored thereon a program that when executed by a computer having a microprocessor causes the computer to implement a method for creating a relativistic metric for measuring distances between sub-structures in an ontology implemented, comprising:

setting, using the microprocessor, a fixed distance, between a top of a first order structure and each sub-structure in the order structure, to a constant value independent of the depth of the order structure, thereby generating relativity between sub-structures, the sub-structures corresponding to concepts, relationships or attributes of an entity;

creating, using the microprocessor, a model of the ontology based on a hierarchy of the sub-structures in the order structure using the fixed distance and a plurality of factors forming the basis of a relativistic conceptual distance metric;

obtaining a data set;

generating, using the microprocessor, a second order structure for the data set;

generating, using the microprocessor, a semantic distance field for clustering by applying the second order structure and the data set to the model of the ontology created in the creating step, the creating step further creating the model of the ontology by mapping concepts and features to create a vector-valued partial order structure based on the plurality of factors forming the basis of the relativistic conceptual distance metric, computing relative conceptual distances using the vector-valued partial order structure, and defining a Semantic Distance Field model based on the relative conceptual distances;

performing data clustering based on the generated semantic distance field in an N dimensional space; and

inducing, using the microprocessor, the ontology from the data clustering performed by the performing step.

12. The non-transitory computer readable medium according to claim 11 , wherein the step of mapping concepts and features to create the vector-values partial order structure includes the step of mapping concepts and features to Rvachev-functions to create the vector-values partial order structure.

13. The non-transitory computer readable medium according to claim 12 , wherein the step of mapping concepts and features to Rvachev-functions to create the vector-values partial order structure, further includes:

generating, for each object of a plurality of objects, equations representing boundaries of attributes for each respective object;

converting, for each object of a plurality of objects, the equations into greater than or equal to zero type inequalities;

generating, for each object of a plurality of objects, a logical expression combining regions of space defined by the inequalities into a semantic entity;

substituting, for each object of a plurality of objects, the logical expression with a corresponding rvachev-function such that the resulting rvachev-function is equal to 0 on a boundary of the semantic entity, greater then 0 inside a region of the semantic entity, and less then 0 outside the region of the semantic entity; and

generating a composite rvachev-function representing logical statements corresponding to the plurality of objects using the respective rvachev-functions of the objects.

14. The non-transitory computer readable medium according to claim 11 , wherein the step of computing relative conceptual distances using the vector-valued partial order structure further includes using a potential function or a virtual gravity function.

15. The non-transitory computer readable medium according to claim 11 , wherein the plurality of factors forming the basis of the relativistic conceptual distance metric include one of a content factor, a relevancy factor, a density Factor, a depth factor, relationship type factor and a directed order factor.