US 11,698,922 B2Granted

Classification and moderation of text

AssigneeValve Corporation

InventorMorten Pedersen

Life of this patent

Estimating the term…

Loading patent…

Abstract

Disclosed herein are techniques and systems for classifying and moderating text using a machine learning approach that is based on a word embedding process. For instance, word embedding vectors may be used to determine clusters of associated text (e.g., similar words) from a corpus of comments maintained by a remote computing system. The remote computing system may then identify, within the corpus of comments, a subset of comments that include text from a given cluster that was determined, from human labeling input, to include a particular type of word or speech. Using this information, the corpus of comments may be labeled with one of multiple class labels. A machine learning model(s) may be trained to classify text as one of the multiple class labels using a sampled set of labeled comments as training data. At runtime, text can be moderated based on its class label.

Claims20

Dependent claims

A system comprising:

one or more processors;
and
memory storing computer-executable instructions that, when executed by the one or more processors, cause the system to:
provide, as input to a trained machine learning model, an unclassified comment that is to be presented on a display of a client machine;
generate, as output from the trained machine learning model, a classification of the unclassified comment to obtain a classified comment;
cause presentation of the classified comment on the display of the client machine with text of the classified comment concealed on the display of the client machine;
receive, from the client machine, an indication that a user of the client machine has requested to view the text of the classified comment;
in response to receiving the indication, cause the text of the classified comment to be temporarily revealed on the display of the client machine;
and
receive, from the client machine, at least one of:
user feedback that is usable to retrain the trained machine learning model;
or
an indication that the user has selected a control element to permanently reveal the text of the classified comment on the display of the client machine.

10 more elements

A method comprising:

providing, as input to a trained machine learning model, unclassified text that is to be presented on a display of a client machine;
generating, as output from the trained machine learning model, a classification of the unclassified text to obtain classified text;
moderating the classified text;
receiving, from the client machine, an indication that a user of the client machine has requested to view the classified text;
in response to the receiving of the indication, causing the classified text to be temporarily revealed on the display of the client machine;
and
receiving, from the client machine, at least one of:
user feedback that is usable to retrain the trained machine learning model;
or
an indication that the user has selected a control element to permanently reveal the classified text on the display of the client machine.

7 more elements

One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by a processor, cause performance of operations comprising:

providing, as input to a trained machine learning model, unclassified text that is to be presented on a display of a client machine;
generating, as output from the trained machine learning model, a classification of the unclassified text to obtain classified text;
moderating the classified text;
receiving, from the client machine, an indication that a user of the client machine has requested to view the classified text;
in response to the receiving of the indication, causing the classified text to be temporarily revealed on the display of the client machine;
and
receiving, from the client machine, at least one of:
user feedback that is usable to retrain the trained machine learning model;
or
an indication that the user has selected a control element to permanently reveal the classified text on the display of the client machine.

7 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/179,781Filed 2 Nov 2018
Published: US 2020/0142999 A17 May 2020
Granted: 11 Jul 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification10 codes

G06F 16/338
Querying: presentation of query results
G06F 16/35
Of unstructured textual data: clustering; Classification
G06F 18/214
Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation: generating training…
G06F 18/2178
Validation; Performance evaluation; Active pattern learning techniques: based on feedback of a supervisor