LEARNING SYNTACTIC PATTERNS FOR AUTOMATIC DISCOVERY OF CAUSAL RELATIONS FROM TEXT

• US 20070282814A1
• Filed: 05/29/2007
• Published: 12/06/2007
• Est. Priority Date: 05/30/2006
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Gupta, Rakesh
• Granted Patent: US 8,244,730 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30

Abstract

The present invention provides a method for extracting relationships between words in textual data. Initially, training relationship data, such as word triplets describing a cause-effect relationship, is received and used to collect additional textual data including the training relationship data. Distributed data collection is used to receive the training data and collect the additional textual data, allowing a broad range of data to be acquired from multiple sources. Syntactic patterns are extracted from the additional textual data and a distributed data source is scanned to extract additional relationship data describing one or more causal relationships using the extracted syntactic patterns. The extracted additional relationship data is then stored, and can be validated by a supervised learning algorithm before storage and used to train a classifier for automatic validation of additional relationship data.

Claims

1. A computer-based method for extracting relationships from textual data, comprising the steps of:

receiving, from a first distributed data source, training data comprising three or more words describing relationships between an action and an object;

collecting textual data including the received training data;

extracting a syntactic pattern from the collected textual data;

scanning a second distributed data source;

extracting target relationships from the second distributed data source using the extracted syntactic pattern; and

storing the target relationships in a computer storage media.

2. The method of claim 1, wherein the step of extracting a syntactic pattern comprises the step of:

generating a dependency tree, the dependency tree describing relationships between words of the textual data.

3. The method of claim 2, wherein the step of extracting a syntactic pattern further comprises the step of:

inserting word order data into the dependency tree.

4. The method of claim 1, wherein the training data comprises one or more triplets of words describing an object, an action and an effect of the action on the object.

5. The method of claim 2, wherein the step of extracting a syntactic pattern further comprises the step of:

inserting satellite links into the dependency tree.

6. The method of claim 1 wherein the step of storing the identified relationships comprises:

determining validity of a target relationship; and

responsive to determining the identified relationship is valid, storing the target relationship.

7. The method of claim 2, wherein the step of extracting a syntactic pattern further comprises preprocessing the textual data to resolve pronouns.

8. A system for extracting relationships from textual data, the system comprising:

an input device for receiving training data and textual data;

a data store, adapted to communicate with the input device, the data store for storing representations of the received training data;

an extraction module, adapted to communicate with the data store and the input device for extracting a syntactic pattern from the textual data using the received training data;

a classifier adapted to communicate with the extraction module and the data store for classifying relationships described by the extracted syntactic pattern.

9. The system of claim 8, wherein the extraction module generates a dependency tree from the textual data:

10. The system of claim 8, wherein the extraction module further preprocesses the text data to identify pronouns in the textual data.

11. The system of claim 9, wherein the extraction module further inserts satellite links into the dependency tree to identify relationships between words in the textual data.

12. The system of claim 8, wherein the extraction module inserts data describing an order of words in the extracted syntactic pattern.

13. A computer program product, comprising a computer readable medium storing computer executable code for extracting relationships from textual data, the computer executable code performing the steps of:

receiving, from a first distributed data source, training data comprising three or more words describing relationships between an action and an object;

collecting textual data including the received training data;

extracting a syntactic pattern from the collected textual data;

scanning a second distributed data source;

extracting target relationship from the second distributed data source using the extracted syntactic pattern; and

storing the target relationships in a computer storage media.

14. The computer program product of claim 19, wherein the step of extracting a syntactic pattern comprises the steps of:

generating a dependency tree, the dependency tree describing relationships between words of the textual data; and

15. The computer program product of claim 18, wherein the step of extracting a syntactic pattern further comprises the steps of:

preprocessing the textual data to resolve pronouns; and

inserting satellite links to the dependency tree.