# US 2014/0207783 A1 · System and method for computerized identification and effective presentation of semantic themes occurring in a set of electronic documents

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

- Number: US 2014/0207783 A1
- Document ID: US20140207783A1
- Status: Published application
- Kind: A1
- Application number: 14/161,221
- Filed: 2014-01-22
- Published: 2014-07-24
- Earliest claimed: 2013-01-22
- Assignees: Microsoft Israel Research and Development (2002) Ltd.
- Applicants: Equivio Ltd.
- Inventors: Yiftach RAVID
- Web page: https://patentgrep.com/patent/US20140207783A1

## CPC

- G06F 16/288: {Entity relationship models}
- G06F 16/285: {Clustering or classification}
- G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

System and method for computerized identification and presentation of semantic themes occurring in a set of electronic documents, comprising performing topic modeling on the set of documents thereby to yield a set of topics and for each topic, a topic-modeling output list of words; and using a processor performing a matching algorithm to match only a subset of each topic-modeling output list of words, to the output list's corresponding topic, such that each word appears in no more than a predetermined number of subsets from among said subsets.

## Claims

### Claim 1

A method for computerized identification and presentation of semantic themes occurring in a set of electronic documents, the method comprising:
performing topic modeling on the set of documents thereby to yield a set of topics and for each topic, a topic-modeling output list of words; and
using a processor performing a matching algorithm to match only a subset of each topic-modeling output list of words, to the output list's corresponding topic, thereby to define respective subsets of words, such that each word appears in no more than a predetermined number of subsets from among said subsets.

### Claim 2 (depends on 1)

A method according to claim 1 wherein said matching algorithm comprises a hospital-residents matching algorithm and each topic's output list of words is used to indicate “hospital” (topic) preferences for “residents” (words).

### Claim 3 (depends on 1)

A method according to claim 1 wherein said predetermined number of subsets comprises one single subset.

### Claim 4 (depends on 1)

A method according to claim 1 wherein said topic modeling comprises LDA-type topic modeling.

### Claim 5 (depends on 1)

A method according to claim 1 wherein each subset has a predetermined maximum number of members such that each subset includes, at most, said maximum number of words.

### Claim 6 (depends on 5)

A method according to claim 5 wherein each subset has a predetermined number of members such that each subset includes exactly said predetermined number of words.

### Claim 7 (depends on 5)

A method according to claim 5 wherein each subset includes less than said maximum number of words if a predetermined computational criterion establishing theme name quality sufficiency of a set of words is satisfied.

### Claim 8 (depends on 2)

A method according to claim 2 wherein said hospital-residents matching algorithm assumes “residents” (words) have no preference between “hospitals” (topics).

### Claim 9

A computer program product, comprising a non-transitory tangible computer readable medium having computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method for computerized identification and presentation of semantic themes occurring in a set of electronic documents, the method comprising:
performing topic modeling on the set of documents thereby to yield a set of topics and for each topic, a topic-modeling output list of words; and
using a processor performing a matching algorithm to match only a subset of each topic-modeling output list of words, to the output list's corresponding topic, such that each word appears in no more than a predetermined number of subsets from among said subsets.

### Claim 10 (depends on 1)

A method according to claim 1 and also comprising displaying said subsets of words as respective theme names for themes corresponding to said set of topics respectively.

### Claim 11

A system for computerized identification and presentation of semantic themes occurring in a set of electronic documents, the system comprising:
topic modeling functionality operative for performing topic modeling on the set of documents thereby to yield a set of topics and for each topic, a topic-modeling output list of words; and
a processor operative for performing a matching algorithm to match only a subset of each topic-modeling output list of words, to the output list's corresponding topic, such that each word appears in no more than a predetermined number of subsets from among said subsets.

### Claim 12 (depends on 11)

A system according to claim 11 wherein said matching algorithm comprises a hospital-residents matching algorithm and each topic's output list of words is used to indicate “hospital” (topic) preferences for “residents” (words).

### Claim 13 (depends on 11)

A system according to claim 11 wherein said predetermined number of subsets comprises one single subset.

### Claim 14 (depends on 11)

A system according to claim 11 wherein said topic modeling comprises LDA-type topic modeling.

### Claim 15 (depends on 11)

A system according to claim 11 wherein each subset has a predetermined maximum number of members such that each subset includes, at most, said maximum number of words.

### Claim 16 (depends on 15)

A system according to claim 15 wherein each subset has a predetermined number of members such that each subset includes exactly said predetermined number of words.

### Claim 17 (depends on 15)

A system according to claim 15 wherein each subset includes less than said maximum number of words if a predetermined computational criterion establishing theme name quality sufficiency of a set of words is satisfied.

### Claim 18 (depends on 12)

A system according to claim 12 wherein said hospital-residents matching algorithm assumes “residents” (words) have no preference between “hospitals” (topics).
