The Spatial Analytics pathway focuses on data analytics techniques dealing with spatial (location) data. Spatial data are increasingly relevant to the goals of many organizations, yet spatial data have unique dimensions that require specialized data models, visualization, and analytic procedures. There is increasing demand for spatially skilled data scientists to exploit the ever-expanding sources of spatial and location-based data.
The courses in this pathway focus on geographic and spatial data, and the robust spatial data management, analysis, and visualization capabilities present in common data science platforms (R and Python). Many analytic capabilities only previously present in Geographic Information Systems (GIS) software have now been migrated to these open source environments where the value of spatial data integrated with non-spatial data sources can be most effectively realized. The pathway is a partnership between UMBC’s Department of Geography and Environmental Science and Department of Computer Science and Electrical Engineering.
Admission into the Data Science graduate program
Proficiency with R and/or Python
GES 773: GIS Project Capstone
This capstone course demonstrates a student’s ability to apply the knowledge and skills attained during their tenure in GIS program. This is a semi-independent course that has students working with industry-relevant clients and undergoing the entire process of developing a real-world solution to meet that client’s needs. These solutions can be in the form of applied research or technological development but will be geared toward the development of some aspect related to an enterprise-wide Geographic Information System. The project will be done in conjunction with a private, government, or academic partner who will be responsible for providing the requirements, data, and system access needed to develop a functional and stable solution in their GIS production environment. Students will produce documentation that demonstrates the design of the GIS, the method for its construction, and instructions for its operation. Guidance from the instructor will interplay with feedback from the client to ensure the student’s success.
GES 774: Statistics for Geographers
The objective of this class is for students to learn how to analyze geographic data using several spatial statistical techniques, grounded in geography principles. Students are given a foundation in basic spatial statistics techniques, including an understanding of how the techniques work in conjunction with the geographic concepts that underpin each. The emphasis in this course is on interpreting and describing analysis results, and less on the statistical mechanics. Students will become familiar with the most common tools used for spatial statistics and gain a detailed understanding of how each technique works. Specifically, students will use several spatial tools in the ArcGIS 10.6 Spatial Statistics Toolbox, CrimeStat 4.0, and Open GeoDa 1.14.
GES 778: Digital Cartography & Visualization
GES 770: Special Topics in Enterprise GIS
This course is to introduce the concept of geoprocessing and how to use Python to automate the process. Students will learn the basic syntax of Python, available system tools in ArcToolbox, model and ModelBuilder. They will also learn how to build a model and how to create a model tool. The core of the class is ArcPy site package. Students will work with the embedded ArcGIS Python window to write and execute scripts. They will also use the more advanced IDE tool Eclipse with PyDev plug-in. Students will write scripts using tools from ArcPy to perform various tasks, such as managing maps and layers, performing data editing, executing geoprocessing tools and creating custom geoprocessing tools. They will develop Python scripts in Eclipse. Once the Python scripts are developed, the student will learn how to convert it to a script tool so it will be available for other Geoprocessing models. In the end of the class, the students will be able to implement a geoprocessing solution for a complex geospatial problem.
GES 775: Advanced Web-GIS Development
The course will review the Python coding basics and NumPy, Pandas to brush on the big data toolkit. It then digs into machine learning and social media toolkit. To transition this course to GIS oriented, it switches the topic to Python GIS toolkit and we will spend time learning mpl_toolkit and BaseMap. Geopandas is a powerful python geospatial tool. Its Geoseries and GeoDataFrame enable convenient data analysis and spatial analysis. Geopandas are unique that they can have multiple geometry columns. You can set its active geometry to display different features. In the same geometry column, it can store different geometry types. Its geometry attributes and methods make spatial analysis straightforward. It inherits all the big data analysis power from pandas. Therefore, all pandas’ capabilities are available in Geopandas. Once we have our focus on GIS, we then move to ArcGIS. It introduces the concept of geoprocessing and how to use Python to automate the process. Students will explore available system tools in ArcToolbox and learn model and ModelBuilder. They will also learn how to build a model and how to create a model tool. One focus of the class is ArcPy site package. Students will work with the embedded ArcGIS Python window to write and execute scripts. They will also use Eclipse with PyDev plug-in. Students will write scripts using tools from ArcPy to perform various tasks, such as managing maps and layers, performing data editing, executing geoprocessing tools, and creating custom geoprocessing tools. They will develop Python scripts in Eclipse. Once the Python scripts are developed, the students will learn how to convert it to a script tool so it will be available for other Geoprocessing models. At the end of the class, the students will be able to implement a geoprocessing solution for a complex geospatial problem.
Spatial Analytics is an emerging field with high job growth. While a subset of the broader discipline of Data Science and Analytics, leveraging location information is fundamental to many corporations (e.g. Uber, UPS, Walmart). Mobile devices, UAV-based sensing and delivery services, and any Location Based Service (LBS) is dependent on spatial data and analytics. According to Labor Insight, an employer-demand tool, employers in the Washington DC and Baltimore metropolitan areas seeking employees with skills in spatial analytics include Booz Allen Hamilton, Vencore, Leidos, and the National Geospatial-Intelligence Agency (NGA).
Download the Academic Planning Form as unofficial guidance in planning your Data Science Master’s program.
Looking for more info?