公开(公告)号：GB2513537A

公开(公告)日：2014-11-05

申请号：GB201223022

申请日：2012-12-20

Applicant: IBM

Inventor： LUKE JAMES STEVEN ,  MAGOWAN JAMES ROBERT ,  BIDDLE EDWARD JAMES ,  WHITE GRAHAM

IPC: G06F17/28

Abstract: A computer implemented method for processing natural language input comprises: receiving natural language input 500; determining whether a rule exists in a rules database for processing the input 505-515, and processing it 520 if this determination is positive; requesting clarification data 535 from a user if this determination is negative; and updating 540 the rules database on the basis of the clarification data. A data processing device for processing natural language input is also provided. The rules for processing may specify actions to be performed, such as store or output, and/or may split up complex input into sub-elements by abstraction 525, and/or may be generalized (470 in Figure 4a) based upon user input. The invention may be used in adding a fact to a fact repository (i.e. fact extraction), querying a fact repository to answer a question, and/or processing a complex sentence.

公开(公告)号：GB2419432A

公开(公告)日：2006-04-26

申请号：GB0423221

申请日：2004-10-20

Applicant: IBM

Inventor： HUNTER ROBERT JAMES ,  LUKE JAMES STEVEN

IPC: G06F40/00 ,  G06N5/04

Abstract: A method for creating hierarchical classifiers of software components in a learning system, each software component comprising an identifier and a rule for processing an input message, the method comprising the steps of: receiving an input message by a first software component; parsing the input message to identify an input value; seeking a second software component having an identifier matching the identified input value; in the event that the seeking step fails to locate a match, creating a second software component and assigning an identifier to the created second software component, the identifier matching the identified input value.