Kekun Hu
Papers
1
Total Citations
41
H-Index
1
About
Kekun Hu is a leading researcher in affective computing and human-robot interaction, with a focus on advancing emotion recognition in conversational AI. His most-cited work, "Hierarchically Stacked Graph Convolution for Emotion Recognition in Conversation" (2023, 41 citations), introduces a novel graph-based framework that captures both self-dependencies and inter-speaker dynamics in dialogue. This approach enables robots to more accurately interpret human emotional cues, paving the way for empathetic and context-aware machine responses. Hu’s contributions lie at the intersection of graph neural networks and affective computing, where he has demonstrated how hierarchical graph structures can model the complex, evolving emotional states in multi-party conversations. His research has significant implications for developing socially intelligent robots capable of nuanced emotional interaction. With a growing citation impact, Hu is recognized for pushing the boundaries of how machines understand and respond to human affect, making his work essential for students and researchers in human-robot interaction, natural language processing, and emotion-aware AI systems.
Research Focus
Key Achievements
Top Papers
- 1