One impact?COVID-19 lockdown has had on us is that it's made some of us very lazy. Even with some of us working from home, we¡¯re finding ways to be more productive, but not compromise on our laziness.?
This laziness also encroaches on our social media accounts, while sharing or commenting on posts. And it turns out, this laziness makes us more human than we could ever imagine being!
Well, don¡¯t take my word for it. This is according to Emilio Ferrara, assistant professor of computer science and the University of Southern California Information Sciences Institute.
His team was studying how the behaviour of humans and bots transitions in a particular session on Twitter.
According to the study published in Frontiers in Physics, If a human is actively commenting on a social media post, with every response he/she will get lazier in his response -- particularly the length of the responses. A bot, however, will be consistent with his levels of engagement.
This drastic difference in behaviour could help in training a new machine learning algorithms for a bot-detection software. Researchers have already added their findings to create a bot detection system that is way ahead of the basic bot detection model.
Today bots on social media have become more prompt and adaptive, making it very difficult to differentiate between human users.?They¡¯re being used to manipulate and influence human users, especially during election campaigns.?
According to Ferrara, ¡°Remarkably, bots continuously improve to mimic more and more of the behaviour humans typically exhibit on social media. Every time we identify a characteristic we think is prerogative of human behaviour, such as sentiment of topics of interest, we soon discover that newly-developed open-source bots can now capture those aspects.¡±
Researchers in the past have focussed on social bot detection, however, they haven¡¯t really given emphasis to measuring the behaviour and activity compared to humans.?The researchers, in order to learn more, used a large Twitter dataset that involved both human accounts as well as bots -- on a post related to recent political events.
The dataset consisted of over 16 million tweets posted by two million different users. The tweets were posted from April 25, 2017 to May 7, 2017 -- a two week period leading to second round of French presidential elections.
They looked at another dataset -- a set of tweets created by bot accounts who were active in viral spamming campaigns, along with a group of human tweets. And in both the data sets they saw a similar trait.
Humans showed a decrease in the amount of content produced -- decreasing average tweet length -- something bots weren¡¯t really doing. Moreover, as the sessions progressed, human users showed fatigue, and engaged in less complex activities.
Research suggests humans would be more likely to get bored by the matter in hand or distracted by something else.
He concluded, ¡°In general, our experiments reveal the presence of a temporal evolution in the human behaviour over the course of a session on an online social network, whereas, confirming our expectations, no evidence is found of a similar evolution for bot accounts. Our analysis highlights the presence of short-term behavioural trends in humans, which can be associated with a cognitive origin, that are absent in bots, intuitively due to the automated nature of their activity.¡±