Predicting Platform Preference of Online Contents Across Social Media Networks

Currently, many professional users tend to promote their websites and brands via multiple online social networks. During activities of information dissemination, the users are confronted with the problem of platform selection. For a post, its platform selection should be based on platform preference, which refers to the platform in which the post can obtain more engagement. In this paper, we focus on this problem by proposing a model to predict platform preference. Specifically, we build a content similarity-based Multi-Task Learning model to predict platform preference of posts. This model takes user specific characters into account and incorporates the regularization term under our validated hypothesis about content similarity. Based on data from Twitter and Facebook, the experiments reveal this model significantly outperforms a number of the baselines. The prediction of platform preference can provide insight for users conducting platform selection to obtain more engagement.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.