Islamabad: Most Pakistanis expressed supportive sentiments towards their Indian neighbours when the latter was reeling under a severe second Covid-19 wave, said an artificial intelligence (AI)-driven study.
The research focused on tweets expressing kindness, empathy and solidarity. It found out that most tweets that were posted by Pakistani citizens between 21 April and 4 May were positive, the Geo TV reported.
Led by Ashique KhudaBukhsh, the team of researchers at Carnegie Mellon University (CMU) based their study on 300,000 tweets. The team accepted only the tweets with the three biggest trending hashtags: #IndiaNeedsOxygen, #PakistanStandsWithIndia and #EndiaSaySorryToKashmir. Of them, 55,712 tweets came from Pakistan, 46,651 from India and the remaining were posted from the rest of the world.
With the help of an artificial intelligence tool, “hope speech classifier”, they found Pakistani tweets containing supportive hashtags were by far more than those containing non-supportive hashtags. The study found that these tweets received more likes and retweets as well.
More than 85 per cent of the tweets posted about the Covid crisis in India from Pakistan were supportive, the research found.
“Our research showed that there’s a universality in how people express emotions. If you search randomly, you’ll find positive tweets a little over 44 per cent of the time. Our method throws up positive tweets 83 per cent of the time,” KhudaBukhsh said.
At a time when Indians were panicked about the raging Covid-19 wave, they received support and solidarity from across the border. Some people justified Pakistan’s helpful outpouring as the country itself was also faced with an infectious outbreak.
“The situation here was pretty bad too. Our hope was getting thinner and thinner. Our enemy was the same, our borders are so close and we get impacted by whatever happens,” said Prof Arifa Zehra, who teaches history in Lahore.
“A pandemic doesn’t recognise borders, whether they are geographical or ideological. And when the dark cloud is sneering at you, there’s no harm in sharing a prayer.”
Prof Zehra also termed these positive tweets as “the greatest reassurance that we are still human”.
KhudaBukhsh expressed hope for better relations between communities and countries if this method of identifying and amplifying positive messages is employed. “When a country is going through a national health crisis like a pandemic, words of hope can be a welcome medicine and the last thing you want to see is negativity,” he highlighted.
“There are several studies that show that if you’re exposed to too much hate speech or negative content, you get influenced by it.”
KhudaBukhsh suggested using this AI-driven method to curb hate speech. “When there’s a negative situation, such as in times of war or a health crisis, instead of blocking the content, an alternative approach can be to highlight the positive content,” he underscored.
“It will help reinforce the belief that people on the other side of the aisle are kinder.”
KhudaBukhsh also suggested building up a robust system that highlights kindness in others before an opposite system is deployed that may censor empathetic content.-IANS
AI-driven tool, what a joke. Man couldn’t they just count and classify terms according to keywords used ?? As the elders say, sophisticated nonsense is still nonsense. It is likely that multiple hashtags were used to trend positive comments as well but the study makes no mention as to how many hashtags were trended in total and how many of them fell into each category i.e positive, negative, neutral, opinionated, etc. In short, the ‘AI-driven study’ makes no effort to determine whether the top-3 hashtags are representative of the total number of tweets made by Pakistanis during the time frame in question (which would be in the millions). The sample was biased as the ‘study’ makes no effort to clarify whether the tweets they had chosen were representative of all the tweets made by Pakistanis during the same period. The so-called ‘study’ can be safely dumped in the junkyard of ‘sophisticated nonsense’. I don’t really expect MM to fully research and understand bias and sample sizes in data science as even people in the industry like me are only beginning to grasp it. I really think MM really needs to hire a technical knowledge reporter or team member, as reading ‘AI’ and ‘hope speech classifier is enough to publish a news article.
@Marquis ke Kalia aka Hindu goo bhakt
You’re trying to hard, troll. The insecure sanghi bhangi is just embarrassing himself. The sample was “biased” and you’re not, goo bhakr? What a clown
Prove me wrong, idiot.
Marquis de kalia, aka OpIndia gaandu
Prove that you’re a liar? I do all the time. Your pedantic and tortured attempts at trying to debunk articles is just reflection of pathetic and low Hindus will stoop to. Babbling about AI and such other nonsense decrypting social media trends. Read your own comments, dickhead…that can be said to cynically “debunk” any social media trends. Lying is natural to the Hindu because your existence is based on myths no person with a single digit IQ takes seriously.
This is why Covid is great for India. No one will miss hate filled Hitler loving Hindus and their primitive idiocy.
You don’t even understand the basics about data collection an processing, so keep out stuff which you have no idea in hell about.
Marquis de kalia aka OPindia troll
Wrong as well. I understand data collection and processing very well. Just like I know how Hindu extremists manipulate information and trends to sway results on your favor. I see it all the time, it’s nothing new. So I know exactly what I’m talking about, and will continue to keep my foot squarely jammed in your smelly lying Hindu ass.
You’re trying too hard in your infantile desperation. At best you’re a low level IT cell computer coolie, at worst just another unemployed troll spamming nonsense for a few rupees. You should upgrade and sell street food.
@Marquis ke Kalia aka Hindu goo bhakt
You’re trying to hard, troll. So now your a statistician in addition to be a RSS goo bhakt?