New artificial intelligence ready to deal with hate speech on social media

Read Time:1 Minute, 34 Second

New artificial intelligence ready to deal with heatspeech on social media — Artificial intelligence system developed by US researchers will highlight positive comments on social media.

New York. The issue is currently debated due to negative and hateful comments circulating on social media. Action is also being taken by the governments of various countries by making rules for this. However, researchers are also now finding scientific ways to deal with it. Currently, scientists have developed an Artificial Intelligence (AI) system that can prevent hate speech against any community.

Developed by researchers from the US-based Carnegie Mellon University, this AI system can rapidly analyze thousands of comments on social media. In addition, it can also recognize comments that are positive.

Researchers said that whether there are minority communities settled in different countries or any particular caste, currently very hateful comments are written against them on social media. However, many comments are in favor or positive. Postdoctoral researcher Ashikul R. Khudabkhsh at the University’s Language Technologies Institute (LTI) said that discovering and highlighting these positive comments can make the Internet as safe and healthy as it is by restricting the search for negative comments and trollers.

Researchers discovered more than one million comments against the particular community on YouTube using this AI system. LTI director and co-author of this study, Gem Carbonate, said that due to recent major improvements in the language model, such a large amount of observations could be analyzed. These models infer what it means for a word to be somewhere. Through this, AI helps in understanding what the writer wants to say.

Researchers have also tested this system in South Asia. He said that it is even more difficult to develop this AI system according to the countries of South Asia, where people speak many languages and write the same statement in different ways. The researchers presented their findings at the annual conference of Artificial Intelligence in New York, US.