반복영역 바로가기
주메뉴로 바로가기
좌측메뉴로 바로가기
본문으로 바로가기

Home PeopleFacultyFaculty

Faculty

프린트페이스북

Jae-Gil Lee

Ph.D. in Computer Science,
KAIST, 2005

Education

  • 1997 KAIST. Computer Science B.S.
  • 1999 KAIST. Computer Science M.S.
  • 2005 KAIST. Computer Science Ph.D.

Research experience

  • 2014. 3 ~ present Associate Professor, Graduate School of Knowledge Service Engineering / Department of Industrial & Systems Engineering, KAIST
  • 2010. 12 ~ 2014. 2 Assistant Professor, Department of Knowledge Service Engineering, KAIST
  • 2008. 9 ~ 2010. 11 Postdoctoral Researcher, IBM Almaden Research Center
  • 2006. 7 ~ 2008. 8 Postdoc Research Associate, Department of Computer Science, University of Illinois at Urbana-Champaign
  • 2005. 3 ~ 2006. 6 Postdoctoral Fellow, Department of Computer Science, KAIST

Research goal

  • Spatio-Temporal Data Mining
  • Social-Network and Graph Data Mining
  • Big Data Analysis with Hadoop/MapReduce and Spark
  • Stream Data Mining and Complex Event Processing
  • High-Performance and Large-Scale Data Warehousing

Research area

Data Mining for Big Data and Data Streams


Big data is everywhere—from the data collected by sensors and to those generated on social networking services. As available data becomes more complex and extensive, weaving it into the knowledge discovery process is a big challenge, with a bigger opportunity for payoff. In this direction, we are extending core data mining algorithms for parallel computing platforms such as Spark and Hadoop to outperform the state-of-the-art technologies (e.g., Mahout and MLlib) and developing a parallel data mining platform for a cluster of mobile devices connected through wireless networks. Furthermore, in order to efficiently process real-time data, we are combining a stream engine (e.g., Storm) and a rule engine (e.g., Esper and Drools) to scale out complex event processing and developing efficient approaches for evaluating continuous queries in a rule engine.

Data Mining for Trajectories, Social Media, and Rich Data Types


The very first issue of data mining and knowledge discovery is to properly handle data. It is essential to take into account different data types. As new services and sensing technologies come about, more complex and various data are being generated in these days. Towards this direction, we are developing novel algorithms and theoretical foundations for data mining of social networks and trajectories as well as the combinations of such data types. The examples include community detection from social networks and temporal group movement patterns from trajectory databases.

Data Mining for Human Behaviors and Mobile Devices


Gartner, Inc. identified “Context-Rich Systems” as one of the top-10 strategic technology trends for 2015. 
Ubiquitous intelligence incorporated into mobile devices will drive the development of systems that are alert to their surroundings and able to respond appropriately. In this direction, we are trying to better understand human behaviors using the data collected by mobile devices such as smartphones and smartwatches. Especially, our current main interests aim towards expertise and interruptibility (the degree of how opportune it is to interrupt a person) in ubiquitous environments.

Selected publications

  • Kim, D., Lee, J., and Lee, B. S., "Topical Influence Modeling via Topic-Level Interests and Interactions on Social Curation Services," In Proc. 32nd Int'l Conf. on Data Engineering (IEEE ICDE), Helsinki, Finland, pp. 13 ~ 24, May 2016.
  • Lim, S., Kim, J., and Lee, J., "BlackHole: Robust Community Detection Inspired by Graph Drawing," In Proc. 32nd Int'l Conf. on Data Engineering (IEEE ICDE), Helsinki, Finland, pp. 25 ~ 36, May 2016.
  • Lee, J., Han, J., and Li, X., "A Unifying Framework of Mining Trajectory Patterns of Various Temporal Tightness," IEEE Trans. on Knowledge and Data Engineering (TKDE), Vol. 27, No. 6, pp. 1478 ~ 1490, June 2015.
  • Lee, J. et al., "Joins on Encoded and Partitioned Data," In Proc. 40th Int'l Conf. on Very Large Data Bases (VLDB) / Proc. of The VLDB Endowment (PVLDB), Vol. 7, No. 13, pp. 1355 ~ 1366.
  • Lim, S., Ryu, S., Kwon, S., Jung, K., and Lee, J., "LinkSCAN*: Overlapping Community Detection Using the Link-Space Transformation," In Proc. 30th Int'l Conf. on Data Engineering (IEEE ICDE), Chicago, Illinois, pp. 292~303, Apr. 2014.
  • Sung, J., Lee, J., and Lee, U., "Booming Up the Long Tails: Discovering Potentially Contributive Users in Community-Based Question Answering Services," In Proc. 7th Int'l AAAI Conf. on Weblogs and Social Media (ICWSM), Cambridge, Massachusetts, pp. 602 ~ 610, July 2013. This paper received the Best Paper Award.
  • Lee, J., Han, J., Li, X., and Gonzalez, H., "TraClass: Trajectory Classification Using Hierarchical Region-Based and Trajectory-Based Clustering," In 34th Int'l Conf. on Very Large Data Bases (VLDB) / Proc. of The VLDB Endowment (PVLDB), Vol. 1, No. 1, pp. 1081 ~ 1094, Aug. 2008.
  • Lee, J., Han, J., and Li, X., "Trajectory Outlier Detection: A Partition-and-Detect Framework," In Proc. 24th Int'l Conf. on Data Engineering (IEEE ICDE), Cancun, Mexico, pp. 140 ~ 149, Apr. 2008.
  • Lee, J., Han, J., and Whang, K., "Trajectory Clustering: A Partition-and-Group Framework," In Proc. 2007 ACM Int'l Conf. on Management of Data (SIGMOD), Beijing, China, pp. 593 ~ 604, June 2007. This paper is the most-cited paper on trajectory data mining. The number of citations is around 900 as of June 2016.

Professional activities

  • Program Chair: 21st Pacific-Asia Conf. on Knowledge Discovery and Data Mining (PAKDD 2017), 6th Database Specialist Workshop (DBSW 2014)
  • Publicity Chair: 17th Int'l Conf. on Database Systems for Advanced Applications (DASFAA 2012), 18th ACM Conf. on Information and Knowledge Management (CIKM 2009)
  • Program Committee Member: KDD 2016, VLDB 2016 (demo track), WAIM 2016; AAAI ICWSM 2015, 2016; SSTD 2015; IEEE ICDE 2015; ACM SIGMOD 2013; ACM CIKM 2009, 2012, 2013, 2016 (demo track); VLDB Ph.D. Workshop 2014; WWW Poster 2014; ER 2012; DASFAA 2008, 2010, 2012, 2014, 2015; APWeb 2010, 2013, 2014; DaWak 2007, 2008; EDB 2012, 2013, 2016; BigComp 2014, 2015, 2016, 2017
  • Local Arrangement Chair: 7th Korea-Japan Database Workshop (KJDB 2012)
  • Information Director: ACM Trans. on Knowledge Discovery from Data (TKDD) (Mar. 2007 ~ Aug. 2008)
  • Board Member: KIISE Database Society of Korea (Aug. 2011 ~ present), Korea Spatial Information Society (June 2015 ~ present)
  • Consultant: Big Data Group, Samsung Software Center (Mar. 2013 ~ Feb. 2015); Knowledge and Information Group, Health Insurance Review & Assessment Service (Apr. 2016 ~ Mar. 2018)
  • Member of ACM and IEEE

Teaching

  • KSE525 Data Mining and Knowledge Discovery
  • KSE526 Analytical Methodologies for Big Data
  • KSE625 Data Mining for Social Networks

Patent

  • Lee, J. and Kim, J., "Device for Detecting Communities and Method for Detecting Communities Using the Same," Korean Patent No: 10-1616477, Apr. 22, 2016.
  • Lee, J. and Sung, J., "Apparatus and Method for Discovering Potentially Contributive Users in Community Based Question Answering Services," Korean Patent No: 10-1563236, Oct. 20, 2015.
  • Lee, J., Sung, J., Kim, A., and Chung, Y., "Method and Apparatus for Managing Tag Based on Trigger," Korean Patent No: 10-1384777-0000, Apr. 7, 2014.
  • Lee, J. and Sung, J., "Terminal, Contents Playback Method of the Same, Message Management System and Contents Message Providing Method of the Same," Korean Patent No: 10-1375791-0000, Mar. 12, 2014.
  • Lee, J., Qiao, L., Raman, V., and Sidle, R., "Multithreaded Data Merging for Multi-Core Processing Unit," U.S. Patent No: US20130138923 A1, May 30, 2013.

LIST