Abstract
With the profusion of web content, researchers have avidly studied and proposed new approaches to enable the anticipation of its impact on social media, presenting many distinct approaches throughout the last decade. Diverse approaches have been presented to tackle the problem of web content popularity prediction, including standard classification and regression approaches. Furthermore, these approaches have also taken into consideration distinct scenarios of data availability, where one may target the prediction of popularity before or after the publication of the items, which is highly interesting for different objectives from a user standpoint. This work aims at reviewing previous work and discussing open issues and challenges that could foster impactful research on this topic. Five areas are identified that require further research, covering the full spectrum of the problem: social media data, the learning task, recommendation and evaluation. © 2019 Elsevier B.V.

Abstract

Abstract

Abstract
The ability to generate and share content on social media platforms has changed the Internet. With the growing rate of content generation, efforts have been directed at making sense of such data. One of the most researched problem concerns predicting web content popularity. We argue that the evolution of state-of-the-art approaches has been optimized towards improving the predictability of average behaviour of data: items with low levels of popularity. We demonstrate this effect using a utility-based framework for evaluating numerical web content popularity prediction tasks, focusing on highly popular items. Additionally, it is demonstrated that gains in predictive and ranking ability of such type of cases can be obtained via naïve approaches, based on strategies to tackle imbalanced domains learning tasks. © 2018 Association for Computing Machinery.

Abstract