# US 2014/0195534 A1 · Creating dimension/topic term subgraphs

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

- Number: US 2014/0195534 A1
- Document ID: US20140195534A1
- Status: Published application
- Kind: A1
- Application number: 13/736,543
- Filed: 2013-01-08
- Published: 2014-07-10
- Earliest claimed: 2013-01-08
- Assignees: International Business Machines Corporation
- Inventors: Judah M. Diament; Aliza R. Heching; Peter K. Malkin
- Web page: https://patentgrep.com/patent/US20140195534A1

## CPC

- G06F 16/35: Clustering; Classification
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A term graph for a group (G), where G is defined by a given set of values d for a set of dimensions (D) relative to a topic (X) may be created by retrieving a graph (H) comprising terms related to an entity and associated with topic X; identifying a node (N) that represents topic X in graph H; identifying resources (R) associated with topic X in group G (used or accessed by, or otherwise associated with values d in group (G); compiling a list (L) of terms used in the identified resources (R); and creating, starting from node N, a connected subgraph S representing the term graph, wherein each node in subgraph S represents one of the terms from list L and has a path to node N.

## Claims

### Claim 1

A method of providing a term graph for a group G, wherein G is defined by a given set of values d for a set of dimensions D relative to a topic X, comprising:
retrieving a graph H comprising terms related to a given entity and associated with the topic X;
identifying a node N that represents the topic X in the graph H;
identifying resources R associated with the topic X and associated with one or more values d of the group G;
compiling, by the processor, a list L of terms used in the identified resources R; and
creating, by the processor, starting from the node N, a connected subgraph S representing the term graph, wherein each node in S represents one of the terms from the list L and has a path to the node N.

### Claim 2 (depends on 1)

The method of claim 1, wherein the resources R comprise one or more of documents, graphics, audio, communications material, or combinations thereof.

### Claim 3 (depends on 1)

The method of claim 1, further comprising persistently storing the given set of values d, the set of dimensions D, the topic X, the graph H, the resources R, the list L, and the connected subgraph S in a database.

### Claim 4 (depends on 1)

The method of claim 1, wherein the dimensions D comprises one or more of particular user, specification of a role, or specification of a time range, or combinations thereof.

### Claim 5 (depends on 1)

The method of claim 1, further comprising obtaining importance measure associated with each of the resources R.

### Claim 6 (depends on 5)

The method of claim 5, further comprising storing the importance measure.

### Claim 7 (depends on 1)

The method of claim 1, further comprising obtaining a distance value of the each node in S from node N.

### Claim 8 (depends on 7)

The method of claim 7, further comprising storing the distance value.

### Claim 9 (depends on 1)

The method of claim 1, wherein the topic X is specified by a text description.

### Claim 10 (depends on 1)

The method of claim 1, wherein two or more term graphs are provided respectively associated with two or more groups G, wherein a node representing a term commonly included in the two or more groups G serves as a connection between the two or more groups G, to create a shared-term term graph.

### Claim 11 (depends on 10)

The method of claim 10, wherein the shared-term term graph is generated automatically in response to a new term graph being added.

### Claim 12 (depends on 10)

The method of claim 10, wherein the shared-term term graph is generated in response to receiving a request to create the shared-term term graph.

### Claim 13 (depends on 10)

The method of claim 10, wherein an importance associated with a term I is a function of an importance associated with the term I in each of the term graphs, the function assigning different weights to said each of the term graphs, wherein the importance associated with the term I indicates a strength of a shared context between the two or more groups G.

### Claim 14 (depends on 1)

The method of claim 1, wherein the node in S representing the term in the list L stores one or more links respectively to one or more of the resources R where the corresponding term in the list L was used.

### Claim 15 (depends on 14)

The method of claim 14, wherein the node in S representing the term in the list L further stores one or more offsets where the term appears in the resources R.

### Claim at position 16

16-25. (canceled)