Based on what we post, the study found there to be largely eight different types of categories of photos on the popular photo sharing app: selfies, food, gadgets, pets, activities, fashion, memes (captioned photos) and posing with friends.
“Our purpose was to try and see if we can understand on a large scale the categories of pictures people posted and then we could figure out the category of users,” said Subbarao Kambhampati, professor of computer science at Arizona State University, where the study was conducted.
The research began with a subset of 95,000 Instagram users.
“After we shortened that to 30,000 random users who were actively posting, we looked at their pictures and categorized,” said Kambhampati, a co-author of the paper. “We then used a computer vision technique that grouped them automatically. It was truly random.”
The study was published in May in the Proceedings of the Eighth International Association for the Advancement of Artificial Intelligence Conference on Weblogs and Social Media, and is believed to be the first of its kind to analyze Instagram photo content and user types.
“As an active Instagram user it makes you think of these things and how people use it,” said Yuheng Hu, a doctoral student at Arizona State University and lead author of the study. “Beyond fashion, food and selfies, what other kind of photos are people posting and what kind of users are there?”
The study also found there to be no direct relationship between the number of followers a person had and the types of pictures they were posting on Instagram. All research excluded "high influence" accounts, such as brands and celebrities.