Sandro Cavallari

Papers

Total Citations

H-Index

About

Sandro Cavallari is a researcher at the forefront of applying graph-based machine learning to social media analytics and user behavior prediction. His work primarily focuses on developing innovative deep learning techniques to model complex, heterogeneous networks, with a particular emphasis on understanding and forecasting user engagement in online video platforms. Cavallari's most notable contribution is his 2021 paper, "Predicting video engagement using heterogeneous DeepWalk," which has garnered 23 citations. In this influential study, he pioneered the use of heterogeneous graph embeddings to capture the intricate relationships between users, videos, and content features, enabling more accurate predictions of video popularity and viewer interaction patterns. This approach represents a significant advancement over traditional homogeneous network models, offering a more nuanced understanding of the dynamics driving online engagement. Cavallari's research is highly relevant for content creators, platform designers, and digital marketers seeking to optimize video performance and user retention. His work continues to shape the integration of graph neural networks into practical, large-scale social media analytics, making him a key figure in the intersection of network science and applied machine learning.