Seungche Kang

Korea Advanced Institute of Science and Technology

Papers

Total Citations

H-Index

About

Seungche Kang is a researcher advancing the frontier of affective computing, with a primary focus on multiclass emotion detection in natural language. His most-cited work, "Detecting Multiclass Emotions from Labeled Movie Scripts" (2018), tackles a critical challenge in AI: moving beyond binary sentiment analysis to precisely discern nuanced negative emotions like anger and sadness—emotions that demand delicate, context-aware responses. By leveraging labeled movie scripts as a rich training corpus, Kang’s research demonstrates how complex emotional states can be systematically extracted from dialogue, offering a pathway toward more emotionally intelligent AI systems. While his citation count (5) reflects a focused, emerging impact, this work stands as a foundational step for researchers exploring fine-grained emotion classification. Kang’s contributions are particularly relevant for applications in human-computer interaction, mental health monitoring, and empathetic dialogue systems, where understanding subtle emotional cues is paramount. His approach highlights the importance of rich, annotated datasets in training models to discern the full spectrum of human affect, marking him as a promising voice in the evolving landscape of AI-driven emotional understanding.