Welcome to Data Mining Laboratory in the Department of Computer Science and Engineering at Seoul National University. Our research interests lie in big data mining which aims to find models, algorithms, and systems for scalable data analysis with applications on knowledge discovery, learning, and anomaly detection. Specifically, we focus on the following research topics: big graph mining, big tensor mining, deep learning & scalable machine learning, online stream mining, anomaly/fraud detection, and recommendation system.
Ph.D. student Jinhong Jung won the HumanTech award (silver, 2nd in Computer Science) from Samsung, from his paper "BePI: Fast and Memory-Efficient Method for Billion-Scale Random Walk with Restart". B.S. student Jungwoo Lee also won the Humantech award (bronze, 3rd in Computer Science), from his paper "CTD: Compact Tensor Decomposition for Multi-way Data". HumanTech award is given to best papers in the engineering area, and it is the most prestigious award among such kind in Korea. Congratulations!
The paper "Fast and Scalable Distributed Boolean Tensor Factorization" is accepted to ICDE 2017, a top tier database conference. The paper proposed a distributed algorithm that performs Boolean tensor factorization in a fast and scalable manner by utilizing caching and minimizing data shuffling.
College of Engineering in Seoul National University selected the 'Seven Technologies to Lead the Future' on the 70th anniversary of foundation. Prof. U Kang introduced Amazon's recommendation system and delivery forecasting system to explain Big Data Technology, one of the seven selected technologies, and predicted the future of real-time big data analysis and forecasting technology.
Prof. Kang won the 1st Young Information Scientist Award from Korean Institute of Information Scientists and Engineers. The award is given to the best Korean Scientist under 40 in the area of Computer Science, and is given to only one person per year. Prof. Kang is the first recipeint of the award. Congratulations!
The paper "BePI: Fast and Memory-Efficient Method for Billion-Scale Random Walk with Restart" is accepted to SIGMOD 2017, a top tier database conference. The paper is on fast and memory-efficient RWR computation in very large graphs exploiting characteristics of real-world graphs.