Location
Havener Center, Miner Lounge / Wiese Atrium, 1:30pm-3:30pm
Start Date
4-1-2026 1:30 PM
End Date
4-1-2026 3:30 PM
Presentation Date
April 1, 2026; 1:30pm-3:30pm
Description
Product quantization has historically been unapplied to GIS datasets, likely due to a mismatch between quantization’s input precondition of fixed, equal length vectors and GIS data’s inherent variability in the number of data points. Therefore, any quantization-compatible encoding method must operate independently of the data points within GIS records. The challenge is to balance the guarantee of producing fixed, equal length vectors with the preservation of semantic meaning present in the raw data. Here, we develop a radial polygon encoding method that achieves this balance while providing acceptable recall in a time complexity of O(nrv), where n is the number of polygons, r is the number of generated rays per polygon, and v is the number of polygon vertices.
Biography
Maris Reinkemeyer is a senior double majoring in Computer Science and Geology & Geophysics. She previously earned an associate’s degree in Computer Application Development from State Technical College of Missouri. She is a member of and women’s academic officer for the Christian Campus Fellowship (CCF), as well as a member of the Honors Academy and a Kummer Vanguard Scholar. She has been working with Dr. Satish Puri since May of 2025, researching product quantization-compatible encoding methods for GIS data. In her free time, she enjoys reading, writing high medieval fantasy, and hanging out with chickens.
Meeting Name
2026 - Miners Solving for Tomorrow Research Conference
Department(s)
Computer Science
Document Type
Poster
Document Version
Final Version
File Type
event
Language(s)
English
Rights
© 2026 The Authors, All rights reserved
Included in
Wheel-spoke encoding: a product quantization-compatible radial encoding scheme for convex polygonal data
Havener Center, Miner Lounge / Wiese Atrium, 1:30pm-3:30pm
Product quantization has historically been unapplied to GIS datasets, likely due to a mismatch between quantization’s input precondition of fixed, equal length vectors and GIS data’s inherent variability in the number of data points. Therefore, any quantization-compatible encoding method must operate independently of the data points within GIS records. The challenge is to balance the guarantee of producing fixed, equal length vectors with the preservation of semantic meaning present in the raw data. Here, we develop a radial polygon encoding method that achieves this balance while providing acceptable recall in a time complexity of O(nrv), where n is the number of polygons, r is the number of generated rays per polygon, and v is the number of polygon vertices.
Comments
Advisor: Satish Puri, satish.puri@mst.edu