Instagram ‘Pods’ Found to be Gaming Algorithm and Boosting Visibility by Coordinated Comments, Likes
Instagram is one of the most popular social media platforms where people get together to share photos and videos. This interaction is quantified by ‘likes' and ‘comments' on posts that are used by the platform to algorithmically highlight popular posts. Now, a team of researchers has found that not all of this engagement is organic. The researchers say there are several groups of users on Instagram who create artificial engagement by liking each other's posts and poses a threat to the integrity of the platform. These groups are popularly known as “pods” or “pod people”.
According to a research conducted by a team at New York University Tandon School of Engineering and Drexel University, there is “a robust underground ecosystem of ‘pods'” on Instagram that artificially boosts content popularity in order to promote their content, increase the reach, or “amplify rhetoric.”
The researchers developed a machine learning tool to uncover the posts on Instagram that have a high likelihood of having gained popularity through pod engagement. Pods also involve humans taking action manually which makes them relatively harder to detect. The style of comments and interaction timings is what allowed the researchers to find posts that were promoted by pod interactions.
Researchers pointed out that it was easy to discover pods via Google search and only 4 percent of the pods discovered required users to have a minimum number of followers before joining.
What was discovered?
According to the researchers, it was found that these pods use “reciprocity abuse” to boost content popularity. Reciprocity abuse is the process where each member, or pod, reciprocally interacts with content posted by other members of the group. This process was found to be very effective as it not only increased visibility of post but, also increased real, organic engagement. They also found that pods are often advertised on “message boards of other pods” that allowed the researchers to discover new pods.
Machine learning was used to analyse 18 lakh Instagram posts that belonged to 1,11,455 unique accounts. These posts were advertised on more than 400 Instagram pods that were hosted on Telegram. The researchers noted that 70 percent of users experienced a two-fold or greater increase in interaction level on control posts and five-fold increase in comments.
“When users who had never posted in pods began posting 50 percent of their posts in pods, they saw a greater than five-fold increase in organic interaction with the posts that they did not post in pods,” the findings state. On average, there were 900 users in a pod, with a maximum of 17,000 users.
What does this mean?
The research shows that pods on Instagram are effective tools for increasing users' Instagram popularity. It states that Telegram-hosted Instagram pods have become quite prevalent and are “likely to continue to be popular.” Their effectiveness poses a threat to the integrity and security of online social networks as they can be used to promote politically motivated propaganda through artificially influence. The researches added that the tool they have created to detect pods “could be deployed as part of content curation algorithms.”
The complete research can be found on ACM Digital Library.
Gloss