# US 2013/0254209 A1 · Consensus search device and method

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

- Number: US 2013/0254209 A1
- Document ID: US20130254209A1
- Status: Published application
- Kind: A1
- Application number: 13/899,786
- Filed: 2013-05-22
- Published: 2013-09-26
- Earliest claimed: 2010-11-22
- Assignees: Korea University Research and Business Foundation
- Applicants: Korea University Research and Business Foundation
- Inventors: Jaewoo Kang
- Web page: https://patentgrep.com/patent/US20130254209A1

## CPC

- G06F 16/2228: {Indexing structures}
- G06F 16/313: {Selection or weighting of terms for indexing}
- G06F 16/3338: {Query expansion}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A consensus search device is provided, which includes a semantic data indexing unit configured to divide text data of an electronic document written about at least one object into segments, to extract at least one semantic descriptor from the each segmented text data, and to generate a semantic data index matching each of the extracted semantic descriptor to the object and the each segmented text data. The consensus search device also includes a semantic searching unit configured to retrieve an object related to a query, based on the semantic data index. The text data is divided into the segments by units of meaning.

## Claims

### Claim 1

A consensus search device comprising:
a semantic data indexing unit configured to divide text data of an electronic document written about at least one object into segments, to extract at least one semantic descriptor from the each segmented text data, and to generate a semantic data index matching each of the extracted semantic descriptor to the object and the each segmented text data; and
a semantic searching unit configured to retrieve an object related to a query, based on the semantic data index,
wherein the text data is divided into the segments by units of meaning.

### Claim 2 (depends on 1)

The consensus search device of claim 1, further comprising:
a metadata indexing unit configured to extract a factual descriptor related to the object from metadata included in the electronic document, and to generate a metadata index matching identification information of the object to the extracted factual descriptor,
wherein the semantic searching unit is configured to retrieve at least one object related to the query by using either or both of the semantic data index and the metadata index.

### Claim 3 (depends on 1)

The consensus search device of claim 1,
wherein the text data is divided into the segments by paragraphs, sentences, clauses, phrases, or words.

### Claim 4 (depends on 3)

The consensus search device of claim 3,
wherein the each segmented text data includes at least one paragraph, at least one sentence, at least one clause, at least one phrase, or at least one word.

### Claim 5 (depends on 1)

The consensus search device of claim 1,
wherein the text data is divided into the segments based on conjunctions or based on parts representing positive sentiment and parts representing negative sentiment.

### Claim 6 (depends on 1)

The consensus search device of claim 1,
wherein the semantic searching unit is configured to calculate the total number of the segmented text data related to the query for each object indicated by the each segmented text data, and to output a search result based on the total.

### Claim 7 (depends on 6)

The consensus search device of claim 6,
wherein the semantic searching unit is configured to generate at least one search keywords by dividing the query into words, to generate extended keywords including at least one of a synonym, a hyponym and a hypernym of the search keywords, and to calculate the total number of the text data of the each segmented text data matched to the semantic descriptors corresponding to at least one of the search keywords and the extended keywords.

### Claim 8 (depends on 7)

The consensus search device of claim 7,
wherein the semantic searching unit is configured to assign different weights to an object matched to the semantic descriptor corresponding to the search keywords and an object matched to the semantic descriptor corresponding to the extended keywords, and to set the weight to the object matched to the semantic descriptor corresponding to the search keywords higher than the weight to the object matched to the semantic descriptor corresponding to the extended keywords.

### Claim 9 (depends on 7)

The consensus search device of claim 7,
wherein when the query includes a word expressing sentiment, the semantic searching unit assigns different weights to objects matched to the semantic descriptors expressing sentiment based on at least one of polarity and degree of the sentiment.

### Claim 10 (depends on 7)

The consensus search device of claim 7,
wherein the semantic searching unit assigns an extra point based on at least one of an author of the segmented text data, the website on which the text data is posted, time point when the text data is posted and information of users' evaluation upon the text data.

### Claim 11 (depends on 7)

The consensus search device of claim 7,
wherein the semantic searching unit generates the search keywords by dividing the query based on units of meaning, and the units of meaning are at least one of sentences, clauses, phrases and words.

### Claim 12

A method for generating an index for consensus search, the method comprising:
(a) dividing text data of an electronic document written about at least one object into segments;
(b) extracting at least one semantic descriptor from the each segmented text data; and
(c) generating a semantic data index matching each of the extracted semantic descriptor to the object and the each segmented text data,
wherein in the process of (a), the text data is divided into the segments by units of meaning.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
(d) extracting a factual descriptor related to the object from metadata included in the electronic document; and
(e) generating a metadata index matching identification information of the object to the extracted factual descriptor.

### Claim 14 (depends on 12)

The method of claim 12,
wherein in the process of (a), the text data is divided into the segments by paragraphs, sentences, clauses, phrases, or words.

### Claim 15 (depends on 12)

The method of claim 12,
wherein the each segmented text data includes at least one paragraph, at least one sentence, at least one clause, at least one phrase, or at least one word.

### Claim 16 (depends on 12)

The method of claim 12,
wherein in the process of (a), the text data is divided into the segments based on conjunctions or based on parts representing positive sentiment and parts representing negative sentiment.

### Claim 17

A consensus search method comprising:
(a) providing a semantic data index by dividing and indexing an electronic document describing at least one object into segments; and
(b) retrieving an object related to a query based on the semantic data index,
wherein the semantic data index is generated by extracting at least one semantic descriptor from the each segmented text data, and matching each of the extracted semantic descriptor to the object and the each segmented text data.

### Claim 18 (depends on 17)

The consensus search method of claim 17,
wherein the process of (b) comprises:
(b1) calculating the total number of the segmented text data related to the query for each object indicated by the each segmented text data;
(b2) calculate degree of relation based on the total; and
(b3) outputting the objects as a search result in the descending order of the degree of relation.

### Claim 19 (depends on 18)

The consensus search method of claim 18,
wherein the process of (b1) comprises:
generating at least one search keywords by dividing the query into words;
generating extended keywords including at least one of a synonym, a hyponym and a hypernym of the search keywords; and
calculating the total number of the text data of the each segmented text data matched to the semantic descriptors corresponding to at least one of the search keywords and the extended keywords.

### Claim 20 (depends on 19)

The consensus search method of claim 19,
wherein the process of (b2) comprises:
calculating assign different weights to an object matched to the semantic descriptor corresponding to the search keywords and an object matched to the semantic descriptor corresponding to the extended keywords, by setting the weight to the object matched to the semantic descriptor corresponding to the search keywords higher than the weight to the object matched to the semantic descriptor corresponding to the extended keywords.

### Claim 21 (depends on 18)

The consensus search method of claim 18,
wherein in the process of (b2), when the query includes a word expressing sentiment, the semantic searching unit assigns different weights to objects matched to the semantic descriptors expressing sentiment based on at least one of polarity and degree of the sentiment.

### Claim 22 (depends on 18)

The consensus search method of claim 18,
wherein in the process of (b2), assigning an extra point based on at least one of an author of the segmented text data, the website on which the text data is posted, time point when the text data is posted and information of users' evaluation upon the text data.
