profile

Yiqun Xie

PhD candidate (started in 2015)
Computer Science and Engineering
University of Minnesota, Twin Cities
Email: xiexx347 at umn.edu

Advisor: Dr. Shashi Shekhar

I am currently a PhD candidate (started in 2015) in Computer Science at the University of Minnesota, Twin-Cities.

I am expected to graduate in Spring/Summer 2020 and seeking a faculty position in the broad area of spatial data science (e.g., data mining, machine learning, optimization and database techniques for spatial data).

NEWS
  • 2019: Our paper has been accepted by ACM Transactions on Data Science (ACM TDS)!
  • 2019: Received the Best Vision Paper Award at the ACM SIGSPATIAL 2019 (Great Innovative Ideas by CCC)!
  • 2019: Our paper has been accepted by the ACM SIGSPATIAL 2019!
  • 2019: Received the Best Paper Award at SSTD 2019!
  • 2019: Our paper has been accepted by ACM Transactions on Intelligent Systems and Technology (ACM TIST)!
  • 2019: Our paper has been accepted by International Symposium on Spatial and Temporal Databases (SSTD'19)!
  • 2019: Our paper has been accepted by International Journal of Geographic Information Science (IJGIS)!
  • 2018: Our paper has been accepted by SIAM International Conference on Data Mining (SDM' 19)!
  • 2018: Our paper has been accepted by IEEE International Conference on Data Mining (ICDM'18)!
  • 2018: Our paper has been accepted by ACM SIGSPATIAL 2018!
Broad research area: Spatial Data Science (spatial data mining, machine learning, optimization, database) with a transdisciplinary view (computer science, statistics, mathematics, domain science)

Sample research projects :
  • Spatial data mining (ACM TDS'20; ACM TIST'19; SSTD'19 Best Paper Award; SDM'19; ICDM'14)
    • Importance and applications: public health (e.g., disease outbreak), public safety, transportation, climate change...
    • Supported by NSF
    • Statistically robust clustering
      • Hotspot detection
      • Significant DBSCAN clustering
    • Significant change interval detection
  • Deep learning for remote sensing data (ICDM'18; ACM SIGSPATIAL'18, 19; IJGIS'18)
    • Importance and applications: Unawareness of the locations of geospatial objects (e.g., individual trees) has caused serious societal problems. For example, a series of deadly fires caused by trees near electricity power lines have killed many people and cost billions of economic loss.
    • Supported by NSF
    • Domain-knowledge guided deep learning for problems with
      • Limited training data
      • Non-distinctive signatures/features
  • Spatial optimization (SSTD'17, AAAI'17 Workshop, ISPRS IJGI'17)
    • Importance and applications: Huge demands on food production (estimated 9 billion population by 2050) vs. severe water pollution (e.g., Dead Zone in the Gulf of Mexico)
    • Optimization with spatially dependent decision variables

Teaching highlights: Courses on machine learning, data science and database.

Service highlights:
Coverage:
Great Innovative Ideas CCC Blog