Which Invention Lets You Know About Social Media Toxicity? | IP Conversations

Social Media is a very important part of life for most of us. but anyone who’s been online for a fair amount of time knows that these spaces can sometimes get very toxic, very fast. But did you know that there’s an invention made specifically to keep track of that?

Read this IP Conversation to find out!

Hey IP Geek!

We all know the internet and social media comes with lot of toxicity. This may be a long shot, but is there any patent to help mitigate that a little?

While controlling social media toxicity is still something people and platforms need to do by themslves, there is a patent which can analyze the possibiliy of a post or conversation becoming toxic.

IBM has patented the invention Social Media Toxicity Analysis in the USA, which claims a method for determining toxicity level and possibiltiy of disengagement in social media conversations.

It does this by first constructing a class model that represents the type of and level of toxicity that is associated with social media content.

When a social media post is analyzed, a “toxicity score” and a class (representing type and topic of content) are assigned to the post. The same is done for the second post in the conversation.

These toxicity scores and classes are then analyzed, and a toxicity trend is forecasted for the conversation thtead, determining how likely it is to go off the rails.

Other factors, such as a “sentiment score” based on the keywords or sentences (determining potential emotions), and the time elapsed between messages in the conversation are also taken into account.

The final toxicity level determined using this method is the probability that the interaction will make someone disengage- what we commonly experience as giving up in rage and frustration.