The Massachusetts Institute Of Technology have released their latest tool, which might just make social media marketing that little bit easier… or more predictable.
We all know that images get more clicks, likes and shares on social media networks such as Facebook and Twitter. The problem comes in knowing exactly what image is most likely to get shared and reach as wide an audience as possible.
M.I.T however have come up with the answer…. possibly. Using a library of 2.3 million images on Flickr, they have designed and coded an algorithym which can predict which image(s) are most likely to provoke a reaction from social media users.
To quote M.I.T directly. “Hundreds of thousands of photographs are uploaded to the internet every minute through various social networking and photo sharing platforms. While some images get millions of views, others are completely ignored. Even from the same users, different photographs receive different number of views. This begs the question: What makes a photograph popular? Can we predict the number of views a photograph will receive even before it is uploaded? These are some of the questions we address in this work. We investigate two key components of an image that affect its popularity, namely the image content and social context. Using a dataset of about 2.3 million images from Flickr, we demonstrate that we can reliably predict the normalized view count of images with a rank correlation of 0.81 using both image content and social cues. In this paper, we show the importance of image cues such as color, gradients, deep learning features and the set of objects present, as well as the importance of various social cues such as number of friends or number of photos uploaded that lead to high or low popularity of images.”
So we put the tool to the test. It’s a long standing marketing joke that if all else fails, use a photo of a cute kitten. So, we used the M.I.T tool to compare a photo of a cute kitten, against a photo of quite possibly the most boring road in the UK, the M6.
For reference, the two original images are available;
So, how did the M.I.T testing tool work. Did it manage to predict which of the two images would have the highest chance of being liked on social media?
So it looks like the M.I.T predictor got this one right. Just.
Seriously speaking though, the full documentation of the study is worth a read, if only the section ‘what is image popularity’. You can read the full document here. The very existence of this topic indicates that imagery and popularity are becoming big business, and knowing which images are most likely to be popular amongst your customers could lead to potential social media gains.