Researchers develop bots that can detect depression based on Twitter profiles

Rate this post
Researchers are developing bots that can detect depression based on Twitter profiles

As social media platforms such as Facebook, Twitter, and Instagram continue to evolve, the recession has been rising among users for many years. Citing this, technology companies and researchers have come up with ways to detect and resolve user mental health problems.Now researchers from Europe have developed Advanced algorithms that can detect depression 9 out of 10 Twitter users. Details are here.

Researchers develop depression detection bot for Twitter

A team of researchers at the University of Brunel and the University of Leicester in London have devised an advanced algorithm that can analyze a user’s mental state based on a Twitter profile. This algorithm takes 38 different data points from a user’s Twitter profile and analyzes them to detect if they are experiencing depression.

Researchers have developed a new algorithm using two databases. One contained the Twitter history of many Twitter users, while the other contained information about mental Used 80% of the data to teach bots And 20% of the data to train it.

When the algorithm works, the bot first excludes users with less than five tweets and runs the rest of the profile in natural language software to check for misspelled words and detect abbreviations.Then it takes into account 38 specific data points, including the use of positive and negative words, emojis and other elements Determine the user’s mental state.

When testing a new depression detection Twitter bot using the Tsinghua Twitter Depression Dataset, Researchers were able to achieve an accuracy of 88.39%, It’s pretty impressive. The bot achieved 70.69% accuracy on the Johns Hopkins University CLPsych 2015 dataset.

“Anything over 90% is considered good at machine learning, so 88% is great in one of the two databases.”Said Professor Abdul Sadka, director of the Institute for Digital Futures at Brunel University. “It’s not 100% accurate, but I don’t think machine learning solutions can achieve 100% reliability at this level. However, the closer you get to the 90% number, the better the results.” The director has added more.

Researchers suggest that new algorithms can be very helpful in detecting mental problems for social media users.According to the team, the bot Can be extended to other platforms such as Instagram and WhatsAppAnd can also be used for future criminal investigations.

“The proposed algorithm is platform-independent and can easily be extended to other social media systems such as Facebook and WhatsApp.” Professor Huiyu Zhou, a professor of machine learning at the University of Leicester, said in a statement. “The next step in this study is to look at its effectiveness in a variety of environments and backgrounds. More importantly, the technologies gained from this study include e-commerce, recruitment screening, candidate screening, and more. It has the potential to develop further into the application of. “ He added further.

So what do you think of the new Twitter bot that detects depression? Let us know your thoughts on it in the comments below, and look forward to more such interesting stories.