Ph.D. candidate in Computer Science

XUN TANG

E-mail: ********@***.*** *******.**@*****.*** Phone: 651-***-****

PROFILE

Xun Tang is a final year Ph.D. candidate in the Department of Computer Science at the University of Minnesota, Twin Cities. His research and algorithm development focuses are on data mining, machine learning, and geographical information science/systems. More specifically, he is interested in developing novel data mining and machine learning techniques for solving critical real world problems from a data-driven perspective. Before starting his Ph.D. education, he received his Master’s and Bachelor’s degrees in computer science from Harbin Institute of Technology, China. Besides data science, his research experience includes computer vision including abnormal behavior detection, object recognition and classification. EDUCATION

• Doctor of Philosophy in Computer Science Aug. 2013 – July 2018 University of Minnesota, Minneapolis, Minnesota (GPA: 3.91)

• Master of Engineering in Computer Science Aug. 2011 – July 2013 Harbin Institute of Technology, Harbin, China

• Bachelor of Engineering in Computer Science Aug. 2007 – July 2011 Harbin Institute of Technology, Harbin, China

TECHNICAL SKILLS

• Programming Languages and Software: Python, Java, Matlab, C/C++, SQL, R

• Data Science: Machine learning, Regression, Classification, Clustering, Anomaly/Outlier detection, Hotspot detection, Network/Graph mining, Spatial Data Science, Trajectory mining

• Statistics: Statistical inference, Hypothesis testing, Spatial statistics, Point process

• Additional Skills: Computer vision, OpenCV programming, Image processing, Data visualization, GIS tools (e.g., QGIS), Parallel programming (e.g., MPI, OpenMP, CUDA), Socket programming, Unix programming

PROJECT EXPERIENCES

• Spatial Hotspots Detection: Developed statistical models and scalable graph mining algorithms for detecting spatial hotspots (i.e. significantly dense clusters) from human activities, traffic accident, on- board vehicle measurement, and trajectory datasets on both two-dimensional geographical area and road networks. Proposed approaches enforce statistical significance (e.g., p-value) on discovered pattern via hypothesis testing on likelihood ratio and achieve dramatically speedup (e.g., 103 times faster) compared to baseline approaches. Implementation is in Java and MATLAB. Related publications include [1, 2, 3].

• Public Health Pattern Discovery: Developed statistical models and algorithms to discover co- location, hotspot, and inequality from cancer, birth defects, environmental and demographic datasets. Proposed approaches extend point process models (e.g., Ripley’s K-function) for aggregate data such as census-block level disease count. Example patterns discovered include interesting co-location between high screening rate and high fatality rate of breast cancer in North Carolina and hotspot of lung cancer along lower Mississippi River. Implementation is in Java, Python, and MATLAB.

• Location-based Service Enhancement via Machine Learning Techniques: Developed a system of algorithms to automatically label “usage rules” (e.g., no smoking) on web maps (e.g., Open Street Map, Apple Maps). Proposed approaches detect and classify “usage rule” signs from large-scale publicly available geo-tagged photo datasets (e.g., Flickr) and map the signs on the effective region with a precision above 90%. Implementation is in Java and MATLAB. Related publications include [4].

• Abnormal Behavior Detection from video: Developed a sparse coding-based motion recognition model to detect abnormal behavior in crowded people from image and surveillance video. The proposed approach applies L1 regularization and uses LASSO to find the optimal solution. It uses bag- of-words for detecting global abnormal pattern and peak-threshold for local abnormal pattern. The proposed approaches outperformed state-of-the-art approaches in terms of area under ROC. Implementation is in C++ with OpenCV library and MATLAB. Related publications include [5, 6]. 2

SELECTED AWARDS & HONORS

• Mentor, Scholar of Distinction in Mathematics and Applied Geography, Wayzata High School

• National Graduate Scholarship, Ministry of Education in China.

• Outstanding Master Thesis, Harbin Institute of Technology.

• Outstanding Undergraduate Thesis, Harbin Institute of Technology.

• Tencent Innovation Scholarship, Tencent, Inc.

• Excellent Student Award, Harbin Institute of Technology.

• First-class Scholarship of Academics, Harbin institute of Technology. SELECTED PUBLICATIONS

1. Xun Tang, Emre Eftelioglu, Dev Oliver, Shashi Shekhar, “Significant Linear Hotspot Detection”, IEEE Transactions on Big Data 3(2), pp. 140-153, 2017.

2. Xun Tang, Emre Eftelioglu, Shashi Shekhar, “Detecting Isodistance Hotspots on Spatial Networks: A Summary of Results”, International Symposium on Spatial and Temporal Databases, pp. 281-299. Springer, Cham, 2017.

3. Xun Tang, Emre Eftelioglu, Shashi Shekhar. “Elliptical Hotspot Detection: a Summary of Results”, In proceedings of 4th International ACM SIGSPATIAL Workshop on Analytics for Big Geospatial Data, pp. 15-24, 2015.

4. Pavel, Samsonov, Xun Tang, Johannes Schoning, Werner Kuhn, Brent Hecht, “You Can’t Smoke Here: Towards Support for Space Usage Rules in Location-aware Technologies”, in Proceedings of 33rd Annual ACM Conference on Human Factors in Computing Systems, pp. 971-974, 2015. 5. Xun Tang, Shengping Zhang, Hongxun Yao, “Sparse Coding based Motion Attention for Abnormal Event Detection”, 20th IEEE International Conference on Image Processing on, pp. 3602-3606, 2013. 6. Xun Tang, “Book Retrieval based on Near-Duplicate Image Matching”, In proceedings of 9th International Conference Fuzzy Systems and Knowledge Discovery, pp. 2616-2619, 2012. TEACHING EXPERIENCE

• Graduate Teaching Assistant, University of Minnesota

– CSCI 2033: Elementary Computational Linear Algebra

– CSCI 5715: From GPS & Virtual Globe to Spatial Computing

– Massive Open Online Course (MOOC) on Coursera.org: From GPS and Google Maps to Spatial Computing

• Graduate Teaching Assistant, Harbin Institute of Technology

– Computer Vision

– Digital Image Processing

Contact this candidate