Iti Chaturvedi

Papers

Total Citations

H-Index

About

Iti Chaturvedi is a researcher whose work lies at the intersection of machine learning, network analysis, and multimedia analytics. Her key research areas include graph-based deep learning, social media behavior prediction, and the application of heterogeneous network embeddings to complex real-world data. Chaturvedi’s most notable contribution is her pioneering approach to predicting user engagement with online video content. In her highly cited 2021 paper, “Predicting video engagement using heterogeneous DeepWalk,” she introduced a novel framework that leverages heterogeneous graph embeddings to capture the intricate relationships between users, videos, and metadata. This work, which has garnered 23 citations, demonstrates how DeepWalk-style random walks can be adapted to non-homogeneous networks, significantly improving the accuracy of engagement forecasts compared to traditional methods. By bridging the gap between graph representation learning and multimedia analytics, Chaturvedi has provided a powerful tool for platforms seeking to understand and optimize content virality. Her research not only advances theoretical understanding of heterogeneous network dynamics but also offers practical insights for recommendation systems and digital marketing. Chaturvedi’s work continues to inspire new directions in predictive modeling for social media and video platforms.