Queensland University of Technology Researcher Team’s Developed Algorithm that can Detect Abusive Contents on Twitter Against Women

Do you know someone who loves tweeting every now and then? Have you ever come across abusive tweets against female users on twitter? Now time has come to behold your fingers from randomly abusing co-twitter female users. 

A researcher team has come up with an algorithm that can detect abusive posts or tweets against women on Twitter easily. Now this is something going to give a new direction to social media platforms- keeping everyone safe online.

A researchers team from Queensland University of Technology (QUT) have developed an algorithm that can now detect abusive and misogynistic content on social media platforms. It can detect from millions of tweets and identify abusive words like slut, whore, rape that directly points out a woman.

We are living in a time where everyone can share their thoughts, views and agree or disagree to others. But some antisocial people are misusing social platforms and targeting women based on caste, colour, religion. 

By sharing harmful or sexual violent content they are targeting women for part of their fun activity. Sometimes the issue gets so serious that it comes under the cyber protection law and the user gets banned. But what about the online image of the woman? 

As precaution is always better than cure- the violence should be stopped at the earlier stage. So, with the latest algorithm from the Queensland University of Technology’s researcher team, women can tweet, retweet and add comments without the fear of getting abused. 

The research team has handled a data set of 1 million tweets and refined them by searching for abusive words in it.

The accuracy of identifying the abusive words by the algorithm is 75 percent and they have a target to make it 100 percent.

The team says that they have executed a deep learning algorithm known as “Long-short term memory with transfer learning”. The machine will look at the previous understanding of terminology and with time it will change the model.

If the system understands the terminology used in the tweet context, it will be easier to detect those abusive words. Data science has improved in such a way that machines can also understand the natural language- it is complicated though.

Tone of statement and meaning of the context are purely dependent on each other and that’s what machines try to understand. Thanks to the machine learning technology- now every activity can be monitored.  

As per computer scientist Richi Nayak, they hope the model can be adopted and proved beneficial for social media sites so that abusive content will be automatically identified and reported. This way women will stay safe over the internet and the social media platform will be a better place to dwell in.

Recent Articles

spot_img

Related Stories

Leave A Reply

Please enter your comment!
Please enter your name here

Stay on top - Get the daily news in your inbox