Applied Spatiotemporal Data Mining应用时空数据挖掘

Course description
With the continuing advances of geographic information science and geospatial
technologies, spatially referenced information have been easily and increasingly
available in the past decades and becoming important information sources in
scientific research and decision making processes. To effectively take advantage
of the rich collection of spatial (and temporal) data, statistical analysis is
often necessary, e.g., to extract implicit knowledge such as spatial relations and
patterns that are not explicit in the data. Spatial data analysis distinguishes
itself from classical data analysis in that spatial analysis focuses on locations,
areas, distances, relationships and interactions of measurements that are usually
referenced as points, lines, and areal units in geographical spaces. In the
past decades, a plethora of theory, methods and tools of spatial analysis have
been developed from different perspectives, and converged as fruitful fields of
geographic information science (GIScience) and spatial statistics.
The purpose of this class is to present the commonly used methods and current
trends in spatial and spatiotemporal data analysis, and innovative applications
in relevant fields (e.g., environmental science and engineering, natural resources
management, ecology, public health, climate sciences, civil engineering, and social
sciences). In this class, we will review the basic principles in spatiotemporal
analysis and modeling, and discuss commonly used methods and tools. Students
are expected to actively participate in class lecture, complete lab assignments,
read assigned articles and develop a project of their own choice or directly related
to their thesis/dissertation topics. The following topics will be covered in the
class, but can be adjustable to meet the students’ interests:
• Exploratory spatial data analysis
• Space-time geostatistics
• Spatial point process and species distribution modeling
• Spatiotemporal disease mapping
• Time series map analysis and change detection
Prerequisites of this course includes an understanding of basic concepts of spatial
analysis and statistics, which could be fulfilled with basic statistics courses or
graduate level of GIS course. Students from different disciplines are welcome,
please contact the instructor should there any question about the prerequisites.
Learning outcomes
After completing this course, the students of this class are expected to be able
• formulate real-world problems in the context of spatial and spatiotemporal
analysis with a knowledge of basic concepts and principles in this field;
• understand commonly used concepts and methods in statistical analysis of
spatiotemporal data;
• apply appropriate spatial and spatiotemporal analytical methods to solve
the formulated problems, and be able to critically review alternative
• utilize programmable scientific computing tools (e.g., R) to make maps,
solve spatial and spatiotemporal analysis problems, and evaluate and assess
the results of alternative methods;
• A reading list of articles will be provided. The following books will be
frequently referred to for reading:
Bivand Roger S., Pebesma, Edzer J., and Gómez-Rubio, Virgilio
(2008), Applied Spatial Data Analysis with R, Springer (eBook available at TTU library).
Cressie, N., & Wikle, C. K. (2011). Statistics for Spatio-temporal
Data. John Wiley & Sons.
Sample course outline
Day Sample topics Readings Hours
1 Class overview and introduction Handouts 3
2 Point pattern analysis Ch.7 BPG 3
3 Species distribution modeling Handouts 3
4 Space-time geostatistics Ch.8 BPG 3
5 Spatiotemporal regression Ch.9,10 BPG 3
6 Time series map analysis Handouts 3
7 Discussion and student presentation 5

Background of Instructor
Dr. Guofeng Cao is an Assistant Professor in the Department of Geosciences
at Texas Tech University. His research interests include geographic information
science and systems (GIS), cyberGIS and spatiotemporal statistics, with a
primary focus on statistical learning of complex spatial and spatiotemporal
patterns across different domains. His research has been supported by different
funding agency. He has published 45 peer-reviewed papers including 30 journal
articles. He received a B.S. in Earth Science from Zhejiang University, an M.S. in
GIS from Chinese Academy of Science, and a M.A. in Statistics and a Ph.D. in
Geography from the University of California, Santa Barbara. He also had several
years of industrial experiences before moving back to academia.