# US 2008/0306944 A1 · Methods, systems and computer program products for analogy detection among entities using reciprocal similarity measures

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2008/0306944 A1
- Document ID: US20080306944A1
- Status: Published application
- Kind: A1
- Application number: 11/035,472
- Filed: 2005-01-14
- Published: 2008-12-11
- Earliest claimed: 2005-01-14
- Assignees: Saffron Technology, Inc.
- Inventors: Manuel Aparicio, IV; Yen-min Huang; David R. Cabana
- Web page: https://patentgrep.com/patent/US20080306944A1

## CPC

- G06F 18/10: Pre-processing; Data cleansing
- G06F 16/215: Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
- G06F 16/2365: {Ensuring data consistency and integrity}
- G06F 16/90339: {by using parallel associative memories or content-addressable memories}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- Y10S 707/99933: Query processing, i.e. searching
- Y10S 707/99936: Pattern matching access
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- Y10S: TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS

## Abstract

Analogies among entities may be detected by obtaining associative counts among the entities and computing similarity measures among given entities and other entities, using the associative counts. First and second entities are then identified as being analogies if the first entity has a strongest similarity measure with respect to the second entity and the second entity also has a strongest similarity measure with respect to the first entity. The similarity measures may be calculated using a normalized entropy inverted among a given entity and other entities.

## Claims

### Claim at position 1

1-3. (canceled)

### Claim 4

A method of detecting analogies among a plurality of entities comprising:
obtaining associative counts among the plurality of entities;
computing respective similarity measures among respective given entities and other entities among the plurality of entities, using the associative counts, by determining a similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity; and
identifying first and second entities from the plurality of entities as being analogies if the first entity has a strongest similarity measure with respect to the second entity and the second entity also has a strongest similarity measure with respect to the first entity,
wherein determining a similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity comprises:
computing a pair entropy of the given entity and the other entity from the perspective of the third entity;
computing an overall entropy of the plurality of entities to one another from the perspective of the third entity;
computing a maximum overall entropy of the plurality of entities to one another from the perspective of the third entity; and
combining the pair entropy of the given entity and the other entity from the perspective of the third entity, the overall entropy of the plurality of entities to one another from the perspective of the third entity and the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity.

### Claim 5 (depends on 4)

A method according to claim 4 wherein the combining comprises dividing the overall entropy of the plurality of entities to one another from the perspective of the third entity minus the pair entropy of the given entity and the other entity from the perspective of the third entity by a log of the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to obtain a result.

### Claim 6 (depends on 5)

A method according to claim 5 wherein the combining further comprises subtracting the result from one.

### Claim at position 5

7-9. (canceled)

### Claim 10 (depends on 4)

A method according to claim 4 wherein computing an overall entropy of the plurality of entities to one another from the perspective of the third entity comprises incrementing at least one variable as new associative counts are obtained so that the overall entropy need not be recomputed as new associative counts are obtained.

### Claim 11

A method of determining a similarity of a given entity to another entity among a plurality of entities comprising:
computing a pair entropy of the given entity and the other entity from a perspective of a third entity;
computing an overall entropy of the plurality of entities to one another from the perspective of the third entity;
computing a maximum overall entropy of the plurality of entities to one another from the perspective of the third entity;
combining the pair entropy of the given entity and the other entity from the perspective of the third entity, the overall entropy of the plurality of entities to one another from the perspective of the third entity and the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to thereby determine the similarity of the given entity to another entity among the plurality of entities; and
storing the pair entropy, the overall entropy, the maximum overall entropy and/or the similarity of the given entity to another entity among the plurality of entities.

### Claim 12 (depends on 11)

A method according to claim 11 wherein the combining comprises dividing the overall entropy of the plurality of entities to one another from the perspective of the third entity minus the pair entropy of the given entity and the other entity from the perspective of the third entity by a log of the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to obtain a result.

### Claim 13 (depends on 12)

A method according to claim 12 wherein the combining further comprises subtracting the result from one.

### Claim 14 (depends on 11)

A method according to claim 11 wherein computing an overall entropy of the plurality of entities to one another from the perspective of the third entity comprises incrementing at least one variable as new associative counts are obtained so that the overall entropy need not be recomputed as new associative counts are obtained.

### Claim at position 11

15-17. (canceled)

### Claim 18

A system of detecting analogies among a plurality of entities comprising:
an integrator memory that is configured to store therein associative counts among the plurality of entities;
a cognitive map that is configured to store therein respective similarity measures among respective given entities and other entities among the plurality of entities, based on the associative counts, where a respective similarity measure is a similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity;
a reciprocal filter that is configured to identify first and second entities from the plurality of entities as being analogies if the first entity has a strongest similarity measure with respect to the second entity and the second entity also has a strongest similarity measure with respect to the first entity; and
a processor that is configured to determine the similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity by performing the following:
computing a pair entropy of the given entity and the other entity from the perspective of the third entity;
computing an overall entropy of the plurality of entities to one another from the perspective of the third entity;
computing a maximum overall entropy of the plurality of entities to one another from the perspective of the third entity; and
combining the pair entropy of the given entity and the other entity from the perspective of the third entity, the overall entropy of the plurality of entities to one another from the perspective of the third entity and the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity.

### Claim 19 (depends on 18)

A system according to claim 18 wherein the combining comprises dividing the overall entropy of the plurality of entities to one another from the perspective of the third entity minus the pair entropy of the given entity and the other entity from the perspective of the third entity by a log of the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to obtain a result.

### Claim 20 (depends on 19)

A system according to claim 19 wherein the combining further comprises subtracting the result from one.

### Claim at position 15

21-25. (canceled)

### Claim 26

A computer program product for detecting analogies among a plurality of entities, the computer program product comprising a computer usable storage medium having computer-readable program code embodied in the medium, the computer-readable program code comprising:
computer-readable program code that is configured to obtain associative counts among the plurality of entities;
computer-readable program code that is configured to compute respective similarity measures among respective given entities and other entities among the plurality of entities, using the associative counts, by determining a similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity; and
computer-readable program code that is configured to identify first and second entities from the plurality of entities as being analogies if the first entity has a strongest similarity measure with respect to the second entity and the second entity also has a strongest similarity measure with respect to the first entity,
wherein the computer-readable program code that is configured to compute respective similarity measures among respective given entities and other entities among the plurality of entities, using the associative counts, by determining a similarity of the given entity to another entity from a perspective of a third entity, relative to an overall similarity of entities to one another from the perspective of the third entity comprises:
computer-readable program code that is configured to compute a pair entropy of the given entity and the other entity from the perspective of the third entity;
computer-readable program code that is configured to compute an overall entropy of the plurality of entities to one another from the perspective of the third entity;
computer-readable program code that is configured to compute a maximum overall entropy of the plurality of entities to one another from the perspective of the third entity; and
computer-readable program code that is configured to combine the pair entropy of the given entity and the other entity from the perspective of the third entity, the overall entropy of the plurality of entities to one another from the perspective of the third entity and the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity.

### Claim 27 (depends on 26)

A computer program product according to claim 26 wherein the computer-readable program code that is configured to combine comprises computer-readable program code that is configured to divide the overall entropy of the plurality of entities to one another from the perspective of the third entity minus the pair entropy of the given entity and the other entity from the perspective of the third entity by a log of the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to obtain a result.

### Claim 28 (depends on 27)

A computer program product according to claim 27 wherein the computer-readable program code that is configured to combine further comprises computer-readable program code that is configured to subtract the result from one.

### Claim at position 19

29-31. (canceled)

### Claim 32 (depends on 26)

A computer program product according to claim 26 wherein the computer-readable program code that is configured to compute an overall entropy of the plurality of entities to one another from the perspective of the third entity comprises computer-readable program code that is configured to increment at least one variable as new associative counts are obtained so that the overall entropy need not be recomputed as new associative counts are obtained.

### Claim 33

A computer program product for determining a similarity of a given entity to another entity among a plurality of entities, the computer program product comprising a computer usable storage medium having computer-readable program code embodied in the medium, the computer-readable program code comprising:
computer-readable program code that is configured to compute a pair entropy of the given entity and the other entity from a perspective of a third entity;
computer-readable program code that is configured to compute an overall entropy of the plurality of entities to one another from the perspective of the third entity;
computer-readable program code that is configured to compute a maximum overall entropy of the plurality of entities to one another from the perspective of the third entity;
computer-readable program code that is configured to combine the pair entropy of the given entity and the other entity from the perspective of the third entity, the overall entropy of the plurality of entities to one another from the perspective of the third entity and the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to thereby determine the similarity of the given entity to another entity among the plurality of entities; and
computer-readable program code that is configured to store the pair entropy, the overall entropy, the maximum overall entropy and/or the similarity of the given entity to another entity among the plurality of entities.

### Claim 34 (depends on 33)

A computer program product according to claim 33 wherein the computer-readable program code that is configured to combine comprises computer-readable program code that is configured to divide the overall entropy of the plurality of entities to one another from the perspective of the third entity minus the pair entropy of the given entity and the other entity from the perspective of the third entity by a log of the maximum overall entropy of the plurality of entities to one another from the perspective of the third entity to obtain a result.

### Claim 35 (depends on 34)

A computer program product according to claim 34 wherein the computer-readable program code that is configured to combine further comprises computer-readable program code that is configured to subtract the result from one.

### Claim 36 (depends on 33)

A computer program product according to claim 33 wherein the computer-readable program code that is configured to compute an overall entropy of the plurality of entities to one another from the perspective of the third entity comprises computer-readable program code that is configured to increment at least one variable as new associative counts are obtained so that the overall entropy need not be recomputed as new associative counts are obtained.
