Building and maintaining information extraction rules

• US 10,296,573 B2
• Filed: 08/31/2016
• Issued: 05/21/2019
• Est. Priority Date: 11/16/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Carreno-Fuentes, Arnaldo, Chiticariu, Laura, Kandogan, Eser, Li, Yunyao, Yang, Huahai
• Primary Examiner(s): Hong, Stephen S
• Assistant Examiner(s): Robinson, Marshon L
• Application Number:
  US15/253,613
• Publication Number:
  US 20160371243A1
• Time in Patent Office: 993 Days
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 3/0486 Drag-and-drop
• CPC Class Codes: G06F 40/117 Tagging; Marking up details of markup languages G06F40/143; Designating a block; Setting of attributes style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154
• CPC Class Codes: G06F 40/123 Storage facilities
• CPC Class Codes: G06F 40/166 Editing, e.g. inserting or deleting

Abstract

Methods and arrangements for managing development of information extraction rules. One or more documents are opened for extraction. An interface is provided to create a label and thereupon label a portion of the document. The created label is stored, and an extractor is developed based on the labeling. A test interface is provided for the extractor, and results of a test conducted through the test interface are displayed. The extractor is exported. In accordance with at least one embodiment, developers are presented with eased automated guidance to write extractors, which thereby reduces an overall manual effort involved in extractor development. Generally, a focused, tutorial-type environment serves as a guide based on previously developed best practices.

Claims

1. A method comprising:

opening one or more documents for extraction;

providing an interface, comprising a query language (AQL) editor, to create a label and thereupon label a portion of the one or more documents using at least one labeled example provided by a user and at least one labeled clue provided by a user, wherein the at least one labeled example label provided by the user comprises text within the one or more documents and wherein the at least one labeled clue comprises a clue of interest indicating why the labeled example is desirable for extraction;

storing the created label;

developing an extractor based on the labeling, wherein the developing an extractor comprises inserting a template statement into the AQL editor, receiving input from the user to complete the inserted template statement, and developing AQL for capturing label examples for extraction in an extraction plan, wherein the developing AQL comprises capturing intent and semantics for the extractor based upon the at least one labeled example;

providing a test interface for the extractor;

displaying results of a test conducted through the test interface; and

exporting the extractor.

2. The method according to claim 1, wherein said providing of an interface comprises providing an interface portion to permit said storing of the created label.

3. The method according to claim 1, wherein said developing comprises creating extraction rules and storing the rules.

4. The method according to claim 3, wherein said providing of an interface comprises providing an interface portion to permit said creating and storing of extraction rules.

5. The method according to claim 4, wherein:

said providing of an interface portion to permit said creating and storing of extraction rules comprises providing an interface for creating at least one top-level node configured to capture at least one main concept to be extracted; and

said creating comprises creating the at least one top-level node.

6. The method according to claim 5, wherein each at least one top-level node comprises examples for recording labeled snippets and extraction rules for capturing labeled snippets.

7. The method according to claim 6, wherein:

said creating of the at least one top-level node comprises creating the examples;

said creating of the examples comprises creating examples of the at least one main concept and creating clues configured to guide extraction of the at least one main concept.

8. The method according to claim 6, wherein the extraction rules for capturing labeled snippets comprise code generation for rules from actions performed with respect to the at least one top-level node.

9. The method according to claim 5, wherein the interface for creating at least one top-level node comprises a drag and drop function that coordinates with a text editor.

10. The method according to claim 1, wherein said providing of a test interface comprises affording a capability to select snippets and clues to label and to display extraction results associatively with respect to labels.

11. The method according to claim 1, wherein the test interface comprises a portion of the interface.

12. The method according to claim 1, further comprising functionally coordinating the test interface with a remainder of the interface such that actions undertaken with respect to the test interface are reflected in the remainder of the interface and actions undertaken with respect to the remainder of the interface are reflected in the test interface.

13. The method according to claim 1, wherein said inserting a template statement comprises conveying a predetermined structure with respect to at least one rule associated with at least one top-level node to thereby guide a user in developing the rules for that node.

14. The method according to claim 1, wherein the template statement comprises at least one member selected from the group consisting of:

basic features, candidate generation, and consolidation.

15. The method according to claim 14, wherein said providing of an interface comprises providing a mechanism for a user to alter the template statement.

16. The method according to claim 1, wherein said providing of an interface to label comprises providing a text editor.

17. An apparatus comprising:

at least one processor; and

a computer readable storage medium having computer readable program code embodied therewith and executable by the at least one processor, the computer readable program code comprising;

computer readable program code configured to open one or more documents for extraction;

computer readable program code configured to provide an interface, comprising a query language (AOL) editor, to create a label and thereupon label a portion of the one or more documents using at least one labeled example provided by a user and at least one labeled clue provided by a user, wherein the at least one labeled example provided by the user comprises text within the one or more documents and wherein the at least one labeled clue comprises includes a clue of interest indicating why the labeled example is desirable for extraction;

computer readable program code configured to store the created label;

computer readable program code configured to develop an extractor based on the labeling, wherein the developing an extractor comprises inserting a template statement into the AQL editor, receiving input from the user to complete the inserted template statement, and developing AQL for capturing label examples for extraction in an extraction plan, wherein the developing AQL comprises capturing intent and semantics for the extractor based upon the at least one labeled example;

computer readable program code configured to provide a test interface for the extractor;

computer readable program code configured to display results of a test conducted through the test interface; and

computer readable program code configured to export the extractor.

18. A computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code configured to open one or more documents for extraction;

computer readable program code configured to provide an interface, comprising a query language (AQL) editor, to create a label and thereupon label a portion of the one or more documents using at least one labeled example provided by a user and at least one labeled clue provided by a user, wherein the at least one labeled example provided by the user comprises text within the one or more documents and wherein the at least one labeled clue comprises a clue of interest indicating why the labeled example is desirable for extraction;

computer readable program code configured to store the created label;

computer readable program code configured to develop an extractor based on the labeling, wherein the developing an extractor comprises inserting a template statement into the AQL editor, receiving input from the user to complete the inserted template statement, and developing AQL for capturing label examples for extraction in an extraction plan, wherein the developing AOL comprises capturing intent and semantics for the extractor based upon the at least one labeled example;

computer readable program code configured to provide a test interface for the extractor;

computer readable program code configured to display results of a test conducted through the test interface; and

computer readable program code configured to export the extractor.

19. The computer program product according to claim 18, wherein said providing of an interface comprises providing an interface portion to permit said storing of the created label.