# US 2010/0169309 A1 · System, Method, and Apparatus for Information Extraction of Textual Documents

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

- Number: US 2010/0169309 A1
- Document ID: US20100169309A1
- Status: Published application
- Kind: A1
- Application number: 12/346,117
- Filed: 2008-12-30
- Published: 2010-07-01
- Earliest claimed: 2008-12-30
- Assignees: Complyon Inc.
- Inventors: Leslie A. Barrett; Morton D. Mackof
- Web page: https://patentgrep.com/patent/US20100169309A1

## CPC

- G06F 16/951: Indexing; Web crawling techniques
- G06F 16/3331: Query processing
- G06F 16/953: Querying, e.g. by the use of web search engines
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method and system for extraction of text from a set of text document(s). A plurality of document segments are identified within a given text document. For each document segment, at least one structured annotation is embedded within the document and associated with the given segment. The structured annotation specifies the start and end of the given document segment and a rhetorical relation associated with the given segment. The structured annotations are processed generate a plurality of variables that represent document segments and associated rhetorical relations as specified by the structured annotations, and such variables are stored in a data repository. A user interacts with a computer to define query input that specifies at least one rhetorical relation of interest. The query input specified by the user is processed to query the variables stored in the data repository to identify zero or more document segments that are associated with a rhetorical relation that matches the at least one rhetorical relation of interest specified by the query input. Information corresponding to the zero or more matching document segments is returned to the user. In the preferred embodiment, the rhetorical relations represented by the user supplied query input as well as the variables stored in the data repository include a set of RST relations whose meaning is dictated by nuclearity of the associated text. Such RST relations can include a plurality of mononuclear RST relations each having a nucleus and a satellite and a plurality of multinuclear RST relations each having a plurality of nucleus. The rhetorical relations represented by the user supplied query input as well as the variables stored in the data repository can also include a set of Speech Act relations whose meaning extends beyond the situational semantics of the associated text.

## Claims

### Claim 1

A method for extraction of text from a set of text documents, the method comprising the steps of:
a) identifying a plurality of document segments within a given text document;
b) for each given document segment identified in a), generating and storing at least one structured annotation embedded within the document and associated with the given segment, the at least one structured annotation specifying the start and end of the given document segment and a rhetorical relation associated with the given segment;
c) processing the structured annotations generated and stored in b) to generate a plurality of variables that represent document segments and associated rhetorical relations as specified by the structured annotations;
d) storing the variables generated in c) in a repository;
e) receiving query input from a user that specifies at least one rhetorical relation of interest; and
f) in response to receipt of said query input, querying the variables stored in the repository to identify zero or more document segments that are associated with a rhetorical relation that matches the at least one rhetorical relation of interest specified by said query input for output to the user.

### Claim 2 (depends on 1)

A method according to claim 1, wherein:
the rhetorical relations include a set of RST relations whose meaning is dictated by nuclearity of the associated text.

### Claim 3 (depends on 2)

A method according to claim 2, wherein:
said set of RST relations includes a plurality of mononuclear RST relations each having a nucleus and a satellite.

### Claim 4 (depends on 2)

A method according to claim 2, wherein:
said set of RST relations include a plurality of multinuclear RST relations each having a plurality of nucleus.

### Claim 5 (depends on 1)

A method according to claim 1, wherein:
the rhetorical relations include a set of Speech Act relations whose meaning extends beyond the situational semantics of the associated text.

### Claim 6 (depends on 1)

A method according to claim 1, wherein:
said at least one structured annotation generated and stored in b) is an XML tag.

### Claim 7 (depends on 1)

A method according to claim 1, wherein:
the repository include first and second sets of variables, said first set of variables representing document segments specified by the structured annotations generated and stored in b), and said second set of variables representing rhetorical relations specified by the structured annotations generated and stored in b) and linked to variables of the first set.

### Claim 8 (depends on 1)

A method according to claim 1, wherein:
the repository stores ancillary data linked to a given text document; the query input received from the user specifies ancillary data of interest; and the querying of f) filters the matched document segments to identify those document segments belonging to a text document linked to ancillary data corresponding to the ancillary data of interest.

### Claim 9 (depends on 1)

A method according to claim 1, wherein:
the repository stores variables representing one of an actor and role and linked to document segments;
the query input received from the user specifies an actor or role of interest; and
the querying of f) filters the matched document segments to identify those document segments linked to variables representing an actor or role corresponding to the actor or role of interest.

### Claim 10 (depends on 1)

A method according to claim 1, wherein:
the query input received from the user specifies additional search terms; and
the querying of f) filters the matched document segments to identify those document segments that satisfy the additional search terms.

### Claim 11 (depends on 10)

A method according to claim 10, wherein:
the additional search terms comprise one or more key word terms.

### Claim 12 (depends on 1)

A method according to claim 1, wherein:
the query input received from the user specifies a goal or need of the user; and the method further comprising analyzing matched document segments in accordance with the goal or need specified by the user and outputting the results of such analysis to the user.

### Claim 13 (depends on 1)

A method according to claim 1, wherein:
the query input received from the user specifies at least one sorting parameter; and
the method further comprises sorting the matched documents in accordance with the at least one sort parameter specified by the query input and outputting the results in order as sorted to the user.

### Claim 14 (depends on 1)

A method according to claim 1, wherein:
the operations of a) and b) are performed by an automated tool or at least in part by a human operator.

### Claim 15 (depends on 1)

A method according to claim 1, further comprising:
presenting output of the querying to a user in a view that presents document segments that are connected to a particular document segment by a relation of interest.

### Claim 16

A system for extraction of text from a set of text documents comprising:
text document annotation means for identifying a plurality of document segments within a given text document and, for each given document segment, generating and storing at least one structured annotation embedded within the document and associated with the given segment, the at least one structured annotation specifying the start and end of the given document segment and a rhetorical relation associated with the given segment;
annotation processing means for processing the structured annotations generated and stored by the text document annotation means to generate a plurality of variables that represent document segments and associated rhetorical relations as specified by the structured annotations;
a repository storing the variables generated by the annotation processing means; user input query means for receiving query input from a user that specifies at least one rhetorical relation of interest; and
query processing logic, operably coupled to the user input query means and the repository, that utilizes said query input to query the variables stored in the repository to identify zero or more document segments that are associated with a rhetorical relation that matches the at least one rhetorical relation of interest specified by said query input for output to the user.

### Claim 17 (depends on 16)

A system according to claim 16, wherein:
the rhetorical relations include a set of RST relations whose meaning is dictated by nuclearity of the associated text.

### Claim 18 (depends on 17)

A system according to claim 17, wherein:
said set of RST relations includes a plurality of mononuclear RST relations each having a nucleus and a satellite.

### Claim 19 (depends on 17)

A system according to claim 17, wherein:
said set of RST relations include a plurality of multinuclear RST relations each having a plurality of nucleus.

### Claim 20 (depends on 16)

A system according to claim 16, wherein:
the rhetorical relations include a set of Speech Act relations whose meaning extends beyond the situational semantics of the associated text.

### Claim 21 (depends on 16)

A system according to claim 16, wherein:
said at least one structured annotation generated and stored by the text document annotation means is an XML tag.

### Claim 22 (depends on 16)

A system according to claim 16, wherein:
the repository include first and second sets of variables, said first set of variables representing document segments specified by the structured annotations generated and stored by the text document annotation means, and said second set of variables representing rhetorical relations specified by the structured annotations generated and stored by the text document annotation means and linked to variables of the first set.

### Claim 23 (depends on 16)

A system according to claim 16, wherein:
the repository stores ancillary data linked to a given text document; the query input received by the user input query means specifies ancillary data of interest; and
the query processing logic filters the matched document segments to identify those document segments belonging to a text document linked to ancillary data corresponding to the ancillary data of interest.

### Claim 24 (depends on 16)

A system according to claim 16, wherein:
the repository stores variables representing one of an actor and role and linked to document segments;
the query input received by the user input query means specifies an actor or role of interest; and
the query processing logic filters the matched document segments to identify those document segments linked to variables representing an actor or role corresponding to the actor or role of interest.

### Claim 25 (depends on 16)

A system according to claim 16, wherein:
the query input received by the user input query means specifies additional search terms; and
the query processing logic filters the matched document segments to identify those document segments that satisfy the additional search terms.

### Claim 26 (depends on 25)

A system according to claim 25, wherein:
the additional search terms comprise one or more key word terms.

### Claim 27 (depends on 16)

A system according to claim 16, wherein:
the query input received from the user specifies a goal or need of the user; and the system further comprises result presentation logic for analyzing matched document segments in accordance with the goal or need specified by the user and outputting the results of such analysis to the user.

### Claim 28 (depends on 16)

A system according to claim 16, wherein:
the query input received from the user specifies at least one sorting parameter; and
the query processing logic sorts the matched documents in accordance with the at least one sort parameter specified by the query input and outputs the results in order as sorted to the user.

### Claim 29 (depends on 16)

A system according to claim 16, wherein:
the text document annotation means comprises an automated tool or carries out operations that are performed at least in part by a human operator.

### Claim 30 (depends on 16)

A system according to claim 16, wherein:
the user input query means and query processing logic are realized by a server coupled to users over a network.

### Claim 31 (depends on 16)

A system according to claim 16, wherein:
the user input query means and query processing logic are realized by a computer processing system accessible by one or more users.

### Claim 32 (depends on 16)

A system according to claim 16, further comprising:
result presentation logic for presenting output of the query processing logic to a user in a view that presents document segments that are connected to a particular document segment by a relation of interest.
