Feeling we have ‘Run out of gas’ emotionally? According to a new survey by SimpleTexting, Twitter tops the list of “most harmful apps”. According to 38.1% of surveys, Twitter was the most trolling platform. This compares with 26.9 percent who say trolling on Facebook is a problem and 14.8 percent on Reddit.
Instead of uniting us, social media is apparently alienating us – so much so that six out of 10 people in the SimpleTexting survey said they were afraid to post about certain topics for fear of negative feedback. 90% of respondents said they have seen or heard racist comments from others on their social networks.
In addition, 86% of respondents said they saw inappropriate content about sexual orientation or gender shared by others on their network. Notably, 46% of respondents admitted to interacting negatively with people online, while 87% said they would follow or block someone they did not like on social media.
Spam is a big problem on Twitter
A second study found that Twitter had a higher toxicity problem. It is suffering from serious spam problems. GlobalData, an international analytics firm, has released its results. It has been found that up to 10 percent of Twitter users are spammers.
Given that Elon Musk is unable to take control of Twitter, his bid for the company is currently stuck because of a dispute over the percentage of spam accounts. Twitter says bot / spam accounts make up less than 5 percent of Twitter accounts, while GlobalData senior data scientist Siddharth K. Kumar says otherwise.
Kumar said it was “difficult to calculate the exact proportion of spam accounts” because it was almost impossible to identify the person behind the tweet handle. The definition of a spam account may vary from person to person. Spam can also be defined as endless tweeting non-original content, although some people may view it as an active user who shares articles / opinions.
GlobalData has developed a mathematical model that estimates spam accounts using a variety of parameters. It then calculates a weighted scoring, which was used to classify the account as “spam” or “non-spam”. GlobalData determines these parameters by looking at the differences in activity between spam accounts and the average Twitter user.
Kumar said that “there were previously published research papers in the media that looked to specific followers to estimate the spam / bot ratio.” The best way to test live streams was because it is more representative of Twitter activity. Because we wanted to make sure spam wasn’t detected, our guess is somewhat conservative. This estimate is only an estimate. It is impossible to determine whether an account is spam or bot.