A Survey On Prevalent Approaches To Predict The Popularity Of Social Content

The destiny of social media images depends upon their popularity: some of the uploaded images/videos get a lot of fame among people while others just get completely unnoticed. Why is this so? This survey paper addresses this question and discusses all the features related to an image/video that are responsible for its popularity among people. Their must be some common features in the images/videos that gets fame, if we get to learn the pattern or features that are responsible for the fame then we can predict if media will get fame or not before actually uploading the media on social media platforms. The prediction of engagement level for an image can be used by companies to improve their marketing strategies. This can be very useful to target the right audience sagaciously, manage the resources efficiently and make strategical decisions. In this survey paper, different prevalent methodologies are discussed that are used by researchers in the past, the techniques they have used to predict the popularity score of an image for a specific number of days and the types of features they have used. This survey paper manifest that the features related to the image context outperforms the features related to image content.