Both readers and online publishers get affected by incendiary comments. Managing them all diligently is a near to impossible task. With too many comments to moderate, the task gets tougher every day. Moreover, for lack of ample workforce to moderate them, only a few comments can be published each-day. Some online publishers are even forced to disable the comments section of their articles and avoid the aftermaths of lewd and unproductive conversations. Apart from outright and unnecessary comments, many are grappling with the menace of incessant comments from programmed bots.
Google and Jigsaw are all set to remove the vitriol out of online comments, and help publishers and media houses address the growing peril successfully. Together they have launched Perspective, an artificial intelligence tool that helps editors identify and manage abusive comments before they get published online. Perspective is an API. It uses machine learning to analyze comments and identify online harassments, insulting conversations, and abuse without the assistance of human moderators. It rates comments based on the scale of toxicity. In other words, “Perspective is an API that makes it easier to host better conversations.”
Most of the times, what’s toxic and what’s not is purely subjective. But Perspective takes a broader approach and decentralizes the power. Thousands of people can judge comments and share the toxicity scale. This helps adopt an unbiased approach to improve online conversations.
The search engine giant has been testing a version of Perspective with The New York Times. It has helped them moderate and enable more comments on each article. For The New York Times team, comment management was both laborious and time consuming during the pre-Perspective days. An entire team had to sift through and moderate every comment before posting them online. But Perspective has transformed the task into an easier and more effective one.
Developers can ask for access to Perspective API and the approval will be given on a rolling basis. Perspective is still a work in progress, but encompasses strong technologies to enhance troll-fighting efforts. Both the Guardian and the Economist are now using API to improve comments sections on their sites. According to Google, increasing usage of the API will help Perspective "develop a better understanding of what makes certain comments toxic”.
Can Perspective turn the tide against bullies, haters, and trolls? Will it work, or join the long list of failed discoveries? Can it identify and handle the complicated nuances within every comment? We’ll have to wait and see!