جستجوی استاد راهنمای دانشگاهی

Kijung Shinمشاهده پروفایل

دانشیار

• Data Mining
• Graph Algorithms
• Network Science
• +۴ مورد دیگر

Kijung Shin is an Associate Professor at KAIST (Korea Advanced Institute of Science and Technology), holding dual appointments in the Kim Jaechul Graduate School of AI and the School of Electrical Engineering (Computer Division). He leads the Data Mining Lab and teaches multiple courses including Graph Mining and Social Network Analysis, Data Mining and Search, and other foundational courses in electrical engineering and AI. Education Ph.D. in Computer Science, Carnegie Mellon University (February 2019) M.S. in Computer Science, Carnegie Mellon University (December 2017) B.S. in Computer Science and Engineering, Seoul National University (August 2015) B.A. in Economics (Double Major), Seoul National University (August 2015) Research Interests Professor Shin's research primarily focuses on data mining, graph algorithms, and network science, with particular expertise in hypergraph analysis, tensor decomposition, and graph neural networks. His work bridges theoretical foundations with practical applications, developing algorithms that can efficiently analyze complex real-world networks. His recent research has expanded into multimodal learning, integration of large language models with graph neural networks, and applications in recommendation systems, satellite imagery analysis, and biological data analysis. His approach combines rigorous mathematical analysis with practical implementation, resulting in numerous open-source software tools that have been widely adopted in both academia and industry. His research has significant implications for social network analysis, fraud detection, recommendation systems, and scientific discovery in various domains. Research Trends Professor Shin's recent publications show a clear trajectory toward more complex network structures, particularly hypergraphs that capture higher-order interactions beyond simple pairwise relationships. His work increasingly integrates traditional graph algorithms with deep learning approaches, especially focusing on how graph neural networks can be improved and made more interpretable. There's also a growing emphasis on practical applications in areas like satellite imagery analysis, medical data, and recommendation systems that address real-world challenges. Scientific Awards Received the PAKDD Best Survey Paper Award for 'Multi-Behavior Recommender Systems: A Survey' (2025) Selected as one of the best short paper candidates of ACM RecSys 2024 (top 7) for 'Revisiting LightGCN' (2024) Selected for oral presentation (2.6% of accepted papers) at AAAI 2024 for 'VITA: 'Carefully Chosen and Weighted Less' Is Better in Medication Recommendation' (2024) Received the IEEE ICDM Best Student Paper Runner-up Award for 'TensorCodec: Compact Lossy Compression of Tensors without Strong Data Assumptions' (2023) Received the SIGKDD Best Research Paper Award and CogX Award for Best Student Paper in AI for 'FRAUDAR: Bounding Graph Fraud in the Face of Camouflage' (2016) Received the Best Senior Thesis Award from Seoul National University (2015) Received the Samsung Humantech Paper Award (1st in Computer Science) (2015) Teaching and Mentoring Professor Shin has taught multiple graduate and undergraduate courses at KAIST since 2019, including Graph Mining and Social Network Analysis, Data Mining and Search, and foundational courses in electrical engineering. He has also co-organized tutorials at major conferences including AAAI, KDD, ICDM, and CIKM on advanced topics in hypergraph neural networks and real-world hypergraph analysis. As the leader of the Data Mining Lab, he mentors numerous graduate students and postdoctoral researchers, fostering a collaborative research environment that has produced significant contributions to the field of data mining and network analysis. Research Leadership Professor Shin leads the Data Mining Lab at KAIST, which focuses on developing novel algorithms for analyzing complex networks and high-dimensional data. The lab has produced numerous influential software tools including D-Cube, M-Zoom, CoreScope, and DenseAlert, which are widely used in both academic research and industry applications. His research group maintains active collaborations with institutions worldwide and has received funding from various sources to support their innovative work in data mining and network analysis.

Edward Yoonjae Choiمشاهده پروفایل

دانشیار

• Machine Learning for Healthcare
• Natural Language Processing
• Multimodal Learning

Edward Yoonjae Choi is an Associate Professor at the Kim Jaechul Graduate School of AI, Korea Advanced Institute of Science and Technology (KAIST), having joined in February 2020 after working at Google Brain and Google Health Research from September 2018. His research focuses on advancing AI applications in healthcare through machine learning, natural language processing, and multimodal systems. His primary research areas include: Machine Learning for Healthcare Natural Language Processing Multimodal Learning Dr. Choi has developed influential models such as the Graph Convolutional Transformer and Doctor AI, which gained recognition from Scientific American, World Economic Forum, and Nvidia's blog for their potential in medical data analysis. His work emphasizes extracting clinical insights from electronic health records and unstructured medical text using deep learning architectures. Analysis of his publications reveals a consistent trend toward integrating heterogeneous data modalities in healthcare AI, with recent work like the Graph Convolutional Transformer (2020) demonstrating novel fusion of graph-based methods and transformer architectures for medical record understanding. His research consistently bridges natural language processing with clinical applications, aiming to improve diagnostic accuracy and treatment personalization through interpretable AI systems.

• Artificial Intelligence
• Machine Learning
• Data Mining
• +۴ مورد دیگر

Bruno Miguel Nunes da Silva is an Adjunct Professor at Polytechnic Institute of Setúbal, where he teaches computer science and conducts research in artificial intelligence and data science. He completed his Doctoral Program in Computer Science from Universidade Nova de Lisboa in 2016. His research interests include: Artificial Intelligence Machine Learning Data Mining Artificial Neural Networks Ubiquitous Computing Data Streams Dr. da Silva's scholarly work focuses on developing innovative methodologies for complex data challenges, particularly in dynamic environments. His research on self-organizing maps for non-stationary data streams represents a significant contribution to the field, addressing the evolving nature of modern data landscapes. His work bridges theoretical computer science with practical applications across various domains. As an active contributor to the academic community, he has served as a reviewer for esteemed journals and conferences, ensuring scholarly rigor. He has also shared his expertise through book chapters, extending his influence beyond traditional journal publications. Dr. da Silva maintains a strong presence in the software development community through GitHub, where he contributes to several open-source projects related to his research interests, including frameworks for neural networks and data visualization tools that demonstrate practical implementations of his theoretical work.