Researchers have demonstrated that Artificial Intelligence algorithms can monitor trolls on online social media. Read on to know more…
In the cyber world, it’s fun to engage other online users in conversations, but the comment sections tend to get out of hand to an upsetting degree. Most of us have experienced the Internet troll where a certain type of personality to thank for an increasingly hostile online environment, replete with vulgarity, insults and sometimes even threats.
Recently, reseachers demonstrated that Artificial Intelligence (AI) algorithms can monitor trolls on online social media. Prevention of online harassment requires the rapid detection of offensive, harassing, and negative social media posts, which in turn requires monitoring online interactions. Current methods to obtain such social media data are either fully automated and not interpretable or rely on a static set of keywords, which can quickly become outdated. It is not recommended to have humans to do this work manually and the humans may also be potentially biased.
But the keyword searching suffers from the speed at which online conversations evolve and new terms come up and old terms change meaning. Hence a keyword that was used sincerely one day may be meant sarcastically used in the next time. To meet this challenge, Maya Srikanth, Anima Anandkumar and the team used GloVe (Global Vectors for Word Representation) model that uses Machine Learning algorithms to discover new and relevant keywords.
GloVe is a word-embedding model, meaning that it represents words in a vector space, where the “distance” between two words is a measure of their linguistic or semantic similarity.
Starting with one keyword, this model can be used to find others that are closely related to that word to reveal clusters of relevant terms that are actually in use.
For example, searching Twitter for uses of “MeToo” in conversations yielded clusters of related hashtags like “SupportSurvivors,” “ImWithHer,” and “NotSilent.” This approach gives researchers a dynamic and ever-evolving keyword set to search.
Researchers at Stanford and Cornell have come up with another potential tool for combatting trolls — early detection. They gained access to user comments hosted by Disqus for the sites Breitbart.com, CNN.com and IGN.com, spanning 18 months from March 2012 through August 2013. The scientists captured data including post content, user activity, community response and moderator actions. They compared messages of users who were never banned to messages of users who were permanently banned, and looked at changes in the banned users’ behavior over their time.
Using the quantifiable results, the researchers were able to develop an algorithm (a set of steps used to solve a problem or perform a task) that used as few as five comments to determine who would be banned in the future with 80 percent accuracy. With 10 posts, the results were 82 percent accurate, but performance peaked around 10 posts. Earlier user posts were better for predicting whether they would get banned later. The team achieved a similar level of accuracy over all three online communities. Post deletion by site moderators turned out to be the most informative activity studied, but all the data in aggregate resulted in better accuracy.