Online Harassment : Jigsaw's "Conversation AI" to the rescue
Updated: Mar 19, 2018
" The Perspective API (demo; API documentation) serves models created by the Conversation AI research initiative to enable platforms, developers, and researchers to explore how ML might help conversations online" - https://conversationai.github.io
This software uses Machine Learning algorithms to identify language of abuse and harassment. To get their algorithm to recognize personal attacks they collected data from Wikipedia talk pages and assembled the largest ever online abuse content. After assembling the data-set the algorithm could determine the probability of a given comment being a personal attack as reliably as moderator would do.
Jigsaw is a technology incubator created by Google(https://jigsaw.google.com/), and now operating as a subsidiary of Alphabet Inc.This company focuses on global real problems and how to use technology to solve them. One of their AI product "Conversation AI" focuses primary on Online Harassment and how one gets abused online
One of its best application is in news companies for example "New York Times" now lets users comment on their front page articles by using the Jigsaw's AI capable products.
Love to hear your comments..Happy blogging!!
A detailed explanation, research info can be found here https://conversationai.github.io/