Our Symposium Workshops will be held both on Thursday June 10 and Friday June 11, 2021. Workshops are FREE for individuals from UCGIS member departments or institutions. For all others, there is a $40 fee per workshop. Not sure if your department or university is a UCGIS member? Here's a current membership list.
Workshop 1: Developing Geospatial Machine Learning Algorithms for Government: An applied workshop in R
- Description: All the recent buzz around "geospatial data science" has led seasoned GIS professionals to ask, "how is this any different from the spatial analytics we have been building for years?'" In this workshop we will learn that the goal of these emerging tools is to optimize the allocation of limited resources across space. This workshop will focus on a single chapter from Dr. Ken Steif's new book, Public Policy Analytics: Code & Context for Data Science in Government. The goal will be to develop a predictive model that helps policymakers target resources where they are most needed. Participants will learn how to measure "exposure'" to geographic risk factors; how to develop predictive features that can capture the spatial process or pattern of a given phenomenon; how to estimate and cross-validate geospatial risk prediction models and then validate the utility of that model through the lens of algorithmic fairness.
- Instructor: Ken Steif, City and Regional Planning, University of Pennsylvania
- Time & Date: Thursday, June 10, 2021, from 11:00 am - 1:00 pm EDT (8:00 am - 10:00 am PDT)
- Location: Virtual. Online platform details to be shared later
- Capacity: 20 people
- Prerequisite Knowledge and Workshop Requirements: Participants should feel comfortable coding spatial analytics in R including the tidyverse and sf packages. It would also be helpful, although not required, to have familiarity with machine learning methods including cross-validation.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise
Workshop 2: Full-stack Geo-visualization 101: How to Make Productive Webmaps (WORKSHOP IS FULL)
- Description: The ability to visualize large amounts of information in real time on interactive maps has never been more important for GI scientists and specialists. Web-based and open-source platforms, such as the mapping library leaflet and the noSQL database MongoDB, are becoming more popular and useful for their real-time functionality. This workshop will serve as an introductory technical training on full-stack geo-visualization. The workshop will include an overview of web-based platforms, mapping libraries, and the backend infrastructure. During this interactive and hands-on session, all participants will have a chance to build a small web map project with a provided data pack. The whole process of programming and debugging will be demonstrated. There will be time for both development and answering questions that participants may have.
- Instructor: Luyu Liu, Ohio State University. Mr. Liu is a PhD student in the Department of Geography in the Ohio State University, where he has been leading and participating in multiple geo-visualization projects since 2017. One of his interactive dashboards was awarded a Regional Sustainability Award by the Mid-Ohio Regional Planning Commission, and another project includes a map gallery focusing on data around Columbus, Ohio.
- Time & Date: Friday June 11, 2021, from 2:00 pm - 4:00 pm EDT (11:00 am - 1:00 pm PDT)
- Capacity: 20 people.
- Location: Virtual. Online platform details to be shared later.
- Prerequisite Knowledge and Workshop Requirements: Participants will be most successful if they are familiar with at least the basics of GIS, programming in general (no specific language or web experience necessary), and some database knowledge (if you wish continue the back-end section of beyond this workshop). Details about the installation of applications for both the first part (front-end) and second part (back-end) will be provided to the participants a few weeks prior to the workshop.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise
Workshop 3: Hour of Cyberinfrastructure for GIScience
- Description: This workshop will introduce participants to the first set of lessons in the online Hour of Cyberinfrastructure for GIScience. We will begin by presenting the objectives and framework of the HCI, including a brief overview of the foundations of Cyberliteracy for GIScience. We will then demonstrate the online materials, practice hands-on exercises, and end with an open discussion on how to incorporate these materials into existing courses at the undergraduate level and elsewhere.
- Instructors: Eric Shook, University of Minnesota; Forrest Bowlick, University of Massachusetts-Amherst; Anand Padmanabhan, University of Illinois; and Karen Kemp, University of Southern California
- Time & Date: Friday June 11, 2021, from 2:00 - 4:00 pm EDT (11:00 am - 1:00 pm PDT).
- Capacity: 30 people
- Location: Virtual. Online platform details to be shared later. The HCI lesson demo will be conducted through your web browser.
- Prerequisite Knowledge and Workshop Requirements: There are no prerequisites for this workshop. The HCI project is designed for instructors and learners who are new to cyberinfrastructure and/or GIScience and the demo will be conducted through your web browser.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise
Workshop 4: Introduction to R for Spatial Data (WORKSHOP IS FULL)
- Description: This 2-hour course will introduce participants with limited or no prior experience with the R programming language to the basics of R and RStudio for both general applications and applications using spatial data. The course will start with an introduction to the RStudio interface and how to implement basic programming commands and methods. We will then explore importing external data into R and using R for data cleaning and manipulation. Finally, we will explore using spatial data in R including an introduction to key packages for spatial data management, how to correctly use CRS and projections with spatial data, and basic data management for spatial data.
- Instructor: Kate Vavra-Musser, University of Southern California.
- Time & Date: Thursday June 10, 2021, from 6:00 pm - 8:00 pm EDT (3:00 pm - 5:00 pm PDT)
- Capacity: 30 people
- Location: Virtual. Online platform details will be announced later.
- Prerequisite Knowledge and Workshop Requirements: No prerequisite knowledge is required for this course. This course is intended for participants with limited or no prior experience with R. Prior experience with other programming languages may be helpful but is also not required. Prior to attending the workshop, please download and install R and RStudio if you do not already have both on your machine. I will begin the workshop under the assumption that everyone has R and RStudio already installed.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise.
Workshop 5: Geoprocessing in R
- Description: This 90-minute course will introduce participants to geoprocessing tools and processes in the R programming language. Some prior experience with R is expected. This course will cover spatial data management, basic spatial processes for vector data (including merges, joins, clips, spatial data transformations, calculating buffers and distances, and other processes), raster calculations, and the basics of interpolation methods in R.
- Instructor: Kate Vavra-Musser, University of Southern California.
- Time & Date: Friday June 11, 2021. 6:00 - 7:30 pm EDT (3:00 - 4:30 pm PDT)
- Capacity: 20 people
- Location: Virtual. Online platform details will be announced later.
- Prerequisite Knowledge and Workshop Requirements: Participants are expected to have some prior experience with R and RStudio as well as some basic programming skills and understanding of programming principles. Participants should have prior experience using and manipulating non-spatial data in R but do not need to have prior experience working with spatial data in R. Prior to attending the workshop, please download and install R and RStudio if you do not already have both on your machine. The workshop under the assumption that everyone has R and RStudio already installed.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise.
Workshop 6: Local Modeling and Multiscale Geographically Weighted Regression
- Description: The overall goal of this workshop is to familiarize participants with the core tenants of local spatial modeling and introduce the necessary concepts and tools to conduct a multiscale analysis of geographic processes. In particular, the focus will be centered on geographically weighted models and the recently developed Multiscale Geographically Weighted Regression (MGWR, https://sgsup.asu.edu/sparc/multiscale-gwr) for examining the relationships between a response variable and a set of explanatory variables and how they potentially vary across space. Participants will be instructed in the primary steps to undertake an MGWR analysis, including data preparation, weighting scheme selection, model calibration, and interpretation of the results through inference and mapping of the associated output. A combination of lectures and hands-on training will be offered during the workshop with examples using the programmatic API and point-and-click software. Completing this workshop will provide participants the knowledge and skills to deploy MGWR for their own research to quantify the scale and heterogeneity of geographic processes.
- Time & Date: Friday June 11, 2021, from 2:00 pm - 4:00 pm EDT (11am - 1pm PDT).
- Instructor: Taylor Oshan, University of Maryland. Dr. Taylor M. Oshan is an Assistant Professor in the Center for Geospatial Information Science within the Department of Geographical Sciences at the University of Maryland, College Park. His research interests are centered on developing spatial analysis methods to investigate spatial and temporal processes with applications in the context of urban health and transportation, as well as building open source tools. In particular, his work has focused on spatial interaction models and multiscale local statistical models.
- Capacity: 20 people
- Location: Virtual. Online platform details to be shared later.
- Prerequisite Knowledge and Workshop Requirements:
- Participants' ability to grasp workshop materials will be greatly increased with a basic knowledge of linear regression and some introductory Python programming. However, we will briefly review linear regression before covering more advanced topics and the modeling software is available through a Python command line API and point-and-click software. Both avenues will be introduced, though some advanced functionality will be demonstrated that is only available programmatically. It is recommended if they download the point-and-click software (https://sgsup.asu.edu/sparc/multiscale-gwr) and install the latest versions of the following python packages - mgwr, pandas,geopandas, pysal, and matplotlib - either locally on their own machine or in a Google Colaboratory notebook. Many of these packages are also typically pre-installed in popular scientific computing python distributions such as Anaconda (https://www.anaconda.com/).
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise
Workshop 7: Redistricting Analysis: Using Free Tools to Rebuild Democracy
- Description: During the current year, states will be using 2020 Census data to draw electoral districts to be used for the coming decade for Congressional and state legislative elections. The redistricting process is contentious, with political parties attempting to control the process to draw lines favorable to their party. Reform groups and citizens have pushed for greater transparency, public input, nonpartisan commissions, and neutral criteria. This workshop will make use of web-based tools and open source software to orient participants toward the redistricting process, data, criteria, and evaluation of redistricting plans. Most of the workshop will be conducted using the built-in data and analytical tools of Dave's Redistricting (https://davesredistricting.org), with additional demonstration using QGIS (free/libre) for participants who wish to conduct custom analyses.
- Instructor: Lee Hachadoorian, Temple University. Lee is a free software advocate with research interests in spatial inequality, urban land value, and electoral redistricting. He works for redistricting reform with Concerned Citizens for Democracy, and has been invited faculty at the Voting Rights Data Institute. Lee is Assistant Director of Temple University's PSM in GIS and PSM in Geospatial Data Science, where he teaches courses in Spatial Database Design, Census Data Analysis, and GIS Programming.
- Time & Date: Thursday, June 10, 2021 from 1:30 - 3:30 pm EDT (10:30 am - 12:30 pm PDT)
- Capacity: 20 people
- Location: Virtual. Online platform details will be announced later.
- Prerequisite Knowledge and Workshop Requirements: No prerequisite knowledge. A laptop with an updated web browser. Prior to the workshop, the instructor will advise the registered participants about what free and open source software they will need to install.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise.
Workshop 8: Urban Streetscape Analyses Using Machine Learning and Geo-tagged Street-level Images
- Description: Urban landscapes are the everyday environment of the majority of the global population. As one of the basic units of the city and a focal point of human activity, streets are among the most critical urban landscape features affecting or reflecting people’s lifestyles as well as their physical, mental, and social well-being. A comprehensive quantification of the streetscape (i.e., features and dynamics) may act as an important utility for those investigating the micro-level urban environment. In the last several years, Google Street View images have been widely used for different applications, such as urban greenery mapping, public health, and urban environmental planning. The geo-tagged street-level images that visually depict the streetscapes at ground-level, from the angle and perspective of pedestrians, provide a novel way to measure and map the urban environment from a more human-centric perspective at the street-level. These efforts have been extended to Treepedia, in which street greenery has been mapped for cities around the world. In this workshop, participants will learn how to use geo-tagged images and machine learning for fine-level urban analyses. Specifically, participants will learn how to choose sites from which to sample, differentiate among various forms of Google Street View images via their metadata, undertake image segmentation using machine learning, and generate urban analytics based on these street-level images.
- Instructor: Xiaojiang Li, Temple University. Xiaojiang is a tenure-track assistant professor at Department of Geography and Urban Studies, Temple University. He was a Postdoctoral Fellow at MIT Senseable City Lab. His research focuses on developing and applying geospatial analyses and data-driven approaches in the domain of urban studies. He has proposed to use Google Street View for urban environmental studies and developed the Treepedia project, which aims to map street greenery for cities around the world. He is also working on using human trace data to study human activities and investigate the connection between urban environments and human activities.
- Time & Date: Thursday, June 10, 2021, from 2:30 - 4:30 pm EDT (11:30 am - 1: 30 pm PDT)
- Capacity: 40 people.
- Location: Virtual. Online platform details to be shared later.
- Prerequisite Knowledge and Workshop Requirements: Participants should have general knowledge about GIS and the basics of Python. We will be using open-source Python modules, including Fiona, Shapely, and other machine learning toolkits.
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise.
Workshop 9: Spatio-temporal analysis of COVID-19 Daily Confirmed Cases
- Description: For the first time in history, we are experiencing a global pandemic and analyzing it as it happens. Because the spread of any infectious disease occurs across space and time, spatial data scientists are at the forefront of the efforts to understand the novel coronavirus's spread and help inform and evaluate mitigation strategies. This workshop will explore data sources for the daily number of confirmed cases, identify county-level trends, and group counties with similar patterns of spread together. Next, we will find clusters of high and low infection rates. Using a series of new GIS-based forecasting methods, we will forecast future patterns of spread. Preliminary data analysis reveals stark inequities in infection rates between racial groups in the US. Racially equitable COVID-19 responses are critical to identifying and supporting the unserved and underserved populations most at risk during this pandemic. GIS can help us support these underserved populations. We will explore techniques for identifying these disparities. Finally, in the workshop's hand-on portion, we will create web-based maps, conduct a visual analysis, and build the skeleton of a story map to communicate our analysis results.
- Instructors: Esri staff: Dr. Este Geraghty, Chief Medical Officer and Health Solutions Director; Dr. Lauren Griffin, Product Specialist - Spatial Analysis; Dr. Kevin Butler, Product Development Engineer; Dr. Aileen Buckley, Senior Product Engineer - Living Atlas.
- Time & Date: Thursday, June 10, 2021, from 4:00 - 6:00 pm EDT (1:00 pm - 3:00 pm PDT)
- Capacity: 25 people.
- Location: Virtual. Online platform details to be shared later.
- Prerequisite Knowledge and Workshop Requirements: TBA
- Fees: this workshop is free for anyone affiliated with a UCGIS member institution, or $40 otherwise.
|
