Machine-learning filters for content moderation and reporting
AssigneeMercari, Inc.
InventorsTakuma YAMAGUCHI, Vishal KASHYAP, Kosuke ARASE
Life of this patent
Estimating the term…
Abstract
Embodiments described herein include moderating listings on an e-commerce site automatically and reporting to moderators. False positive reporting may be rejected using algorithms and filtering of reporting may be handled. Methods for moderating terms of service (ToS) violations may include receiving indications of a ToS violation, generating values corresponding to the ToS violation based on a machine learning (ML) model and the indications of the ToS violation. The vales may be evaluated to determine an actual ToS violation or false positives. The ML model may be updated based on the values and the indications of a ToS violation.
Claims20
A computer-implemented method for moderating terms of service (ToS) violations, the computer-implemented method comprising:
- receiving, via at least one computer processor, one or more indications of a ToS violation;
- generating a plurality of values corresponding to the one or more indications of the ToS violation, via the at least one computer processor performing at least one machine-learning (ML) process based at least in part on at least one ML model and the one or more indications of the ToS violation, wherein the plurality of values comprises a classification value and a priority score;
- evaluating, via the at least one computer processor, the plurality of values to yield a first result indicating that the one or more indications of the ToS violation correspond to an actual ToS violation;
- receiving, via the at least one computer processor, a second result indicating that the first result is a false positive, wherein the one or more indications of the ToS violation do not correspond to the actual ToS violation;
- and
- updating, via the at least one computer processor, the ML model, responsive to the second result, based at least in part on the second result and the one or more indications of the ToS violation corresponding to the second result.
A non-transitory computer readable storage medium storing instructions for moderating terms of service (ToS) violations, that, when executed by at least one computer processor, cause the at least one computer processor to perform operations comprising:
- receiving one or more indications of a ToS violation;
- generating a plurality of values corresponding to the one or more indications of the ToS violation, via at least one machine-learning (ML) process based at least in part on at least one ML model and the one or more indications of the ToS violation, wherein the plurality of values comprises a classification value and a priority score;
- evaluating the plurality of values, to yield a first result indicating that the one or more indications of the ToS violation correspond to an actual ToS violation;
- receiving a second result indicating that the first result is a false positive, wherein the one or more indications of the ToS violation do not correspond to the actual ToS violation;
- and
- updating the ML model, responsive to the second result, based at least in part on the second result and the one or more indications of the ToS violation corresponding to the second result.
A system for moderating terms of service (ToS) violations, comprising:
- a memory;
- and
- at least one computer processor coupled to the memory and configured to perform operations comprising:
- receiving one or more indications of a ToS violation;
- generating a plurality of values corresponding to the one or more indications of the ToS violation, via at least one machine-learning (ML) process based at least in part on at least one ML model and the one or more indications of the ToS violation, wherein the plurality of values comprises a classification value and a priority score;
- evaluating the plurality of values, to yield a first result indicating that the one or more indications of the ToS violation correspond to an actual ToS violation;
- receiving a second result indicating that the first result is a false positive, wherein the one or more indications of the ToS violation do not correspond to the actual ToS violation;
- and
- updating the ML model, responsive to the second result, based at least in part on the second result and the one or more indications of the ToS violation corresponding to the second result.
Dependencies are inferred from the claim text.