Back to results
US 9,792,095 B2Granted
Apparatus and method for converting a procedure manual to an automated program
AssigneeSymbol Technologies, Inc.
InventorsYan-Ming M. Cheng, Robert E Beach
Life of this patent
Estimating the term…
Abstract
A computing device receives a query about at least one concept and at least one document associated with the at least one concept. The computing device accesses a plurality of models created based on information in a labeled database. The computing device decodes information in the at least one document using the plurality of models. Responsive to the decoding, the computing device generate a program with steps associated with the at least one concept.
Claims14
A method, comprising:
- receiving, by a computing device, a query about at least one concept, including one or more associated steps, and at least one natural language document associated with the at least one concept;
- accessing, by the computing device, a plurality of models created based on information in a labeled database;
- decoding, by the computing device, information in the at least one natural language document using the plurality of models;
- and
- responsive to the decoding, generating, by the computing device, a program having computer executable code corresponding to the one or more steps associated with the at least one concept,
- wherein:
- (a) the plurality of models include:
- a concept model for determining a specific location in the at least one natural language document that includes the at least one concept;
- a template model for determining at least one of steps associated with the at least one concept and a sub-concept associated with the at least one concept;
- a description model for generating attributes and determining whether a first step depends on another step for execution;
- and
- a sequencing model for determining a sequence of the steps associated with the at least one concept;
- (b) the decoding comprises:
- determining a location of the at least one concept in the at least one natural language document, a location of a sub-concept associated with the at least one concept, and a location of steps associated with at least one of the at least one concept and the sub-concept;
- and
- decoding the at least one concept included in the query using the concept model, the template model, the description model, and the sequencing model.
13 more elements
A computing device, comprises:
- a memory configured to store non-transitory computer-executable instructions;
- a transceiver configured to receive a query about at least one concept, including one or more associated steps, and at least one natural language document associated with the at least one concept;
- a processor configured to perform a set of functions including:
- accessing a plurality of models created based on information in a labeled database;
- decoding information in the at least one document using the plurality of models;
- and
- responsive to the decoding, generating a program having the computer-executable instructions corresponding to the one or more steps associated with the at least one concept,
- wherein:
- (a) the plurality of models include:
- a concept model for determining a specific location in the at least one natural language document that includes the at least one concept;
- a template model for determining at least one of steps associated with the at least one concept and a sub-concept associated with the at least one concept;
- a description model for generating attributes and determining whether a first step depends on another step for execution;
- and
- a sequencing model for determining a sequence of the steps associated with the at least one concept;
- (b) the processor is further configured to perform the decoding by:
- determining a location of the at least one concept in the at least one natural language document, a location of a sub-concept associated with the at least one concept, and a location of steps associated with at least one of the at least one concept and the sub-concept;
- and
- decoding the at least one concept included in the query using the concept model, the template model, the description model, and the sequencing model.
15 more elements
Dependencies are inferred from the claim text.