AI for Content Moderation at PayPal
- Room:
- Liffey Hall 1
- Start (Dublin time):
- Start (your time):
- Duration:
- 45 minutes
Abstract
Online platforms have a hard time combating hate, hate speech, explicit content and other NSFW material. Most of the solutions are rule based keyword approaches which are brittle and can be bypassed easily. At PayPal, we have a wide range of user generated content and there is a great need to automatically identify and flag hate, explicit and other typologies, to improve user experience and adhere to regulatory policies. In this talk we showcase how AI can help us identify such content with great precision.
TalkPyData: Machine Learning, Stats
Description
Online content moderation at scale is a non trivial task especially with an ever changing landscape of hate, hate speech with changing geopolitical scenarios. Moderation platforms need to support multiple typologies like - hate, sexually explicit, violence, bullying, spam and other toxic material. Add multi-language support for all typologies and it becomes an uphill task. In this talk we will cover the below topics:
- Why is Text Content Moderation is hard? Why we need AI?
- What are the available open-source datasets to train models?
- What are the available pre-trained models for content moderation?
- Why pre-trained models do not always work?
- Data labelling strategies and how to leverage open data and models?
- How to build multi-language support and challenges?