Predicting social media post popularity using multimodal deep learning: Insights from content features
Yi Wang, Aoyu Zeng · 2026
Read Predicting social media post popularity using multimodal deep learning: Insights from content features by Yi Wang, Aoyu Zeng (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0356612.txt.
Description
Yi Wang and Aoyu Zeng built deep learning models that predict how popular a social media post will become. The models judge from the post itself: its images and its words. Their 2026 study extracted visual and textual features from real platform data. It confirmed which features track engagement. It then trained two models. One estimates a popularity score, and one sorts posts into four popularity tiers.
Both models performed usefully on the data examined. Combining image and text beat either alone. That supports the idea that posts succeed as packages. The authors also frame popularity as a staged process of spreading through audiences. For marketers, the practical pitch is pre-publication testing. Score a draft post, adjust its content, and spend promotion money where the model sees real potential.
Catalog record
| CREATOR | Yi Wang, Aoyu Zeng |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56799 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |