Back to results
US 10,788,900 B1Granted
Pictorial symbol prediction
AssigneeSnap Inc.
InventorsWilliam Brendel, Francesco Barbieri, Xin Chen, Wei Chu, Venkata Satya Pradeep Karuturi, Luis Carlos Dos Santos Marujo, Leonardo Ribas Machado das Neves
Life of this patent
Estimating the term…
Abstract
Symbol prediction can be implemented using a multi-task system trained for different tasks. The tasks may include a single symbol prediction, symbol category prediction, and symbol subcategory prediction. Categories of symbols can be generated by clustering sets of training data using a clustering scheme.
Claims16
A method comprising:
- identifying one or more text characters input into a client device;
- generating a plurality of pictorial symbol classifications from the one or more text characters using a machine learning scheme, the plurality of pictorial symbol classifications including at least one pictorial symbol category classification for a group of pictorial symbols of a similar type;
- determining that one of the plurality of pictorial symbol classifications satisfies a specified threshold;
- and
- displaying, on a display device of the client device, a presentation of one or more display elements that are pre-associated with the one of the plurality of pictorial symbol classifications that satisfies the specified threshold, wherein the machine learning scheme is a multitask neural network having a first network portion configured to generate individual pictorial symbol classifications and a second network portion configured to generate pictorial symbol category classifications.
2 more elements
A method comprising:
- identifying one or more text characters input into a client device;
- generating a plurality of pictorial symbol classifications from the one or more text characters using a machine learning scheme, the plurality of pictorial symbol classifications including at least one pictorial symbol category classification for a group of pictorial symbols of a similar type;
- determining that one of the plurality of pictorial symbol classifications satisfies a specified threshold;
- and
- displaying, on a display device of the client device, a presentation of one or more display elements that are pre-associated with the one of the plurality of pictorial symbol classifications that satisfies the specified threshold;
- generating one or more images on the client device;
- receiving the one or more text characters as a caption for overlay on the one or more images;
- and
- generating a symbol prediction using a random forest learning scheme, the random forest learning scheme trained on multimodal messages comprising voice data.
6 more elements
A system comprising:
- one or more processors of a machine;
- and
- a memory storing instructions that, when executed by at least one processor among the one or more processors, causes the machine to perform operations comprising:
- identifying one or more text characters input into a client device;
- generating a plurality of pictorial symbol classifications from the one or more text characters using a machine learning scheme, the plurality of pictorial symbol classifications including at least one pictorial symbol category classification for a group of pictorial symbols of a similar type;
- determining that one of the plurality of pictorial symbol classifications satisfies a specified threshold;
- and
- displaying, on a display device of the client device, a presentation of one or more display elements that are pre-associated with the one of the plurality of pictorial symbol classifications that satisfies the specified threshold, wherein the machine learning scheme is a multitask neural network having a first network portion configured to generate individual pictorial symbol classifications and a second network portion configured to generate pictorial symbol category classifications.
5 more elements
A machine-readable storage device embodying instructions that, when executed by a device, cause the device to perform operations comprising:
- identifying one or more text characters input into a client device;
- generating a plurality of pictorial symbol classifications from the one or more text characters using a machine learning scheme, the plurality of pictorial symbol classifications including at least one pictorial symbol category classification for a group of pictorial symbols of a similar type,
- determining that one of the plurality of pictorial symbol classifications satisfies a specified threshold;
- and
- displaying, on a display device of the client device, a presentation of one or more display elements that are pre-associated with the one of the plurality of pictorial symbol classifications that satisfies the specified threshold, wherein the machine learning scheme is a multitask neural network having a first network portion configured to generate individual pictorial symbol classifications and a second network portion configured to generate pictorial symbol category classifications.
2 more elements
Dependencies are inferred from the claim text.