Back to results
US 12,217,002 B2Granted
Parallel processing of hierarchical text
AssigneeNVIDIA Corporation
InventorsElias Stehle, Gregory Michael Kimball
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques to parse textual data using parallel computing devices. In at least one embodiment, text is parsed by a plurality of parallel processing units using a finite state machine and logical stack to convert the text to a tree data structure. Data is extracted from the tree by the plurality of parallel processors and stored.
Claims23
A processor, comprising:
- a plurality of processing units to perform one or more operations from a list of operations comprising:
- grouping one or more sequences of an input stream into one or more tokens based, at least in part, on simulating a finite state transducer;
- identifying, using at least one logical stack, one or more hierarchical relationships between the one or more tokens;
- generating a data tree based, at least in part, on the identified one or more hierarchical relationships;
- and
- analyzing the data tree in parallel to identify one or more shared paths of one or more nodes in the data tree, infer schema information from the one or more shared paths, and store data from the input stream in accordance with the schema information.
3 more elements
A method, comprising:
- grouping one or more sequences of an input stream into one or more tokens based, at least in part, on simulation of one or more finite state machines;
- identifying one or more hierarchical relationships between the one or more tokens, using at least one logical stack;
- generating a data tree based, at least in part, on the one or more identified hierarchical relationships;
- identifying one or more shared paths of one or more nodes in the data tree;
- inferring schema information from the shared paths;
- and
- storing data from the input stream in accordance with the schema information.
4 more elements
A non-transitory computer-readable storage medium having stored thereon instructions that, when executed by at least one processor of a computing device, cause the computing device to at least:
- group sequences of an input stream into tokens based, at least in part, on simulation of a finite state machine;
- identify parent-child relationships between the tokens, using at least one logical stack;
- generate a tree based, at least in part, on the identified parent-child relationships;
- identify shared paths of nodes in the tree;
- infer schema information from the shared paths;
- and
- store data from the input stream in accordance with the schema information.
4 more elements
Dependencies are inferred from the claim text.