Earth Day at KDD
Calling all earth and environmental scientists, students and researchers! Please join us at Earth Day on August 5th!
Why at SIGKDD? Why now?
Earth is home to all of us. However, our beloved planet and civilization is facing major challenges from climate change and environmental degradation. Extreme events are becoming more extreme and more frequent. Surface water has more pollution and greenhouse gases have increased in the atmosphere. Moreover, the largest freshwater source on Earth, polar ice-caps and glaciers are losing ice, leading to sea-level rise.
Knowledge discovery and data mining (KDD) is crucial to addressing these and other challenges facing our changing planet. For example, Earth data (i.e. geo-referenced data such as in-situ and remotely sensed Earth observations, census, trajectories) records and helps in understanding biological, physical, and social changes. It can help forecast rates of sea level change in polar ice shelves, predict critical atmosphere and geospace events. It is also important for many societal priorities including security, public health, smart cities, transportation, climate, environment, food, energy, water.
Earth data have unique characteristics that bring challenges to data science. For some applications, the boundaries of geospatial objects can be amorphous and dynamically deform through time. In addition, the training of machine learning models also faces great challenges due to a lack of training samples: (1) significant Earth events can be very rare (e.g., earthquakes, cyclones), and (2) ground truth data are labor-intensive and time-consuming to collect. Earth Day will bring together thought leaders in academia, industry and government to explore this area and discuss opportunities to overcome the challenges that Earth faces today.
Earth Day Session Program
Morning: Three earth-related workshops
12:00-1:00 pm- Lunch
1:30 pm-1:45 pm: Welcome, Rationale for Earth Day theme
1:45 pm-3:00 pm: Session 1: Importance of Earth data sets and use cases
Background: Earth data (e.g., remote sensing imagery, GPS time service, location traces) has already transformed our lives by improving monitoring of global weather and agriculture for early warning of hurricanes and inclement weather as well as food shortage risks due to crop stresses or failures. Further, with 2 billion  receivers in use for location and time services, the GPS has become a critical infrastructure for the world economy for use cases ranging from precision agriculture to navigation to ride sharing to smart cities. These success stories are only a beginning and many transformative opportunities lie ahead. For example, the 2011 Mckinsey Big Data report  estimated that location trace data will generate about $600 billion annually by 2020. In addition, 2019 U.S. national academy report projects $1.6 trillion in savings for energy generation and use from earth data by 2035 . Furthermore, government and industry have recently started major initiatives such as NASA Earth Exchange, Amazon’s Earth on AWS, Google Earth Engine, Microsoft’s AI for Earth, and NSF Navigating the New Arctic for meeting grand challenges facing our changing planet such as conservation, climate change, and environmental sustainability. This session will explore the tremendous value of Earth data for civil society, prosperity and good governance via a keynote and a panel discussion.
 Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space: An Overview for Decision Makers and the Public. Washington DC: The National Academies Press, 2019.
 “The World Economy Runs on GPS. It Needs a Backup Plan.” Bloomberg, July 2018.
 S. Shekhar et al., Spatial Computing, Communications of the ACM, 59(1):72-81, 2016.
 Personal location data, chapter 3e, in “Big data: The next frontier for innovation, competition, and productivity,” pp. 85-96, McKinsey Global Institute, 2011.Keynote (30 min)
Panel (45 min): Societal value of Earth Data (Chair: James Hodson, AI for Good Foundation)
- Jennifer Marsman, Microsoft AI for Earth/li>
- Dennis Pamlin, RISE Research Institutes of Sweden Holding AB/li>
- To be determined
Questions for Panelists:
- 1a. Civil Society View: What are their societal significance & most important use cases?
- 1b. Business View: What will the annual value of Earth Data be in 2030 or 2040? Why?
- 1c. Governance View: What is the role of Earth data in good governance?
- 2. List important types, sources and accessible repositories of Earth data.
- 3. List unique characteristics of Earth data and unique needs of its use cases.
- 4. List strengths and weaknesses of current data mining techniques for Earth Data.
4:00 pm - 5:15 pm: Session 2 : Earth Day related Data Mining challenges and opportunities
Background:Data mining methods have found success in analyzing many complicated systems, such as e-commerce, and use cases explored in the Earth Day aligned SIGKDD workshops. However, many questions remain open due to unique Earth data challenges such as spatio-temporal auto-correlation, heterogeneity, scale-dependence, measurement errors, modifiable areal unit problem, etc [2,3,5]. For example, a recent Geo-Physical letter paper  noted that “failure to account for dependence between [Physical] models, variables, locations and seasons yield misleading results”. Additional challenges are noted in recent community papers from the NSF IS-GEO Research Coordination Network  and University Consortium for Geographic Information Science . For example, Gerrymandering court debates  also raise transparency concerns for the risk of altering statistical results by changing the choice of spatial partitions. This session explores these challenges and opportunities via a keynote and a panel discussion.
 P. M. Caldwell et al. Statistical significance of climate sensitivity predictors obtained by data mining, Geophys. Res. Lett., 41:1803-1808, AGU, 2014.
 Y. Gil et al. Intelligent systems for geosciences: an essential research agenda. Communications of the ACM, 62(1): 76-84, 2018.
 A UCGIS Call to Action: Bringing the Geospatial Perspective to Data Science Degrees and Curricula, University Consortium for Geographic Information Science, Summer 2018.
 M. Hutson, Has artificial intelligence become alchemy?, Science, 360(6388):478, AAAS, 04 May 2018, (DOI: 10.1126/science.360.6388.478).
 Y. Xie et al. Transdisciplinary Foundations of Geospatial Data Science, ISPRS Int. J. Geo-Inf. 2017, 6, 395.
 Justices Display Divisions in New Cases on Voting Maps Warped by Politics, New York Times. 2019.
- Prof. Dan Griffith, AAAS Fellow
- Prof. Anuj Karpatne, Virginia Tech
- Prof. Tanya Berger-Wolf, University of Illinois at Chicago
Questions for panelists:
- List knowledge gaps between Earth data mining needs and data mining state of the art.
- What new research is needed to fill the knowledge gaps?
- What are the data mining grand challenges with respect to analyzing Earth data?
- Is there a need for SIGKDD community action? If so, suggest community actions.
Shashi Shekhar, University of Minnesota – Twin Cities (Co-chair)
James Hodson, AI for Good Foundation (Co-chair)
Lucas Joppa, Microsoft Research (Co-chair)
Chaitanya Baru, University of California, San Diego
Vandana Janeja, University of Maryland, Baltimore County
Hui Xiong, Rutgers University & Baidu (Beijing, China)
Jieping Ye, University of Michigan, Ann Arbor & DiDi Chuxing (China)
Xun Zhou, University of Iowa
Ramasamy Uthurusamy, General Motors Corporation
Chid Apte, IBM T. J. Watson Research Center
Naoki Abe, IBM T. J. Watson Research Center
Vani Mandava, Microsoft Research
Meredith Lee, West Big Data Hub
Melissa Cragin, San Diego Supercomputing Center (formerly Midwest Big Data Hub)
Lea Shanley, South Big Data Hub
How can we assist you?
We'll be updating the website as information becomes available. If you have a question that requires immediate attention, please feel free to contact us. Thank you!
Please enter the word you see in the image below: