A Topological Approach to Understanding Location-Based Data

Location-based services have seen a boon in data production recently which has simultaneously stoked the research community to better understand this type of information. Traditional methods in analyzing such data require significant a priori understanding of the organization of the data. We submit...

Full description

Saved in:
Bibliographic Details
Main Authors: McAbee, Carson, Wakefield, Max D., Roth, John D., Scrofani, James W.
Format: Conference Proceeding
Language:English
Subjects:
Artificial intelligence
Data models
Machine Learning
Maintenance engineering
Markov processes
Shape
Topological Data Analysis
Training
Training data
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Location-based services have seen a boon in data production recently which has simultaneously stoked the research community to better understand this type of information. Traditional methods in analyzing such data require significant a priori understanding of the organization of the data. We submit that the nascent field of topological data analysis (TDA) may be able to contribute new insights to analysis of such data without the aforementioned requirement. To this end, we propose two novel methods of embedding such data in order to leverage the expressive power of TDA. To demonstrate its effectiveness we apply the embeddings to maritime automated information system data.
ISSN:2576-2303
DOI:10.1109/ACSSC.2018.8645120