Location-Awareness in Time Series Compression

Xu, Teng; Züfle, Andreas; Trajcevski, Goce; Klabjan, Diego

doi:10.1007/978-3-319-98398-1_6

Cited by 3 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much work has been done in terms of investigating different representation methods and distance functions for evaluating similarity of time series [14], and the impact of compression in such settings [6]. The trade-offs between different compression techniques and distance functions in terms of similarity of location-bound time series have been investigated in [12].…”

Section: Introductionmentioning

confidence: 99%

Cet-Lats

Giri

Hashemi

Gossling

et al. 2020

Proceedings of the 28th International Conference on Advances in Geographic Information Systems

Self Cite

View full text Add to dashboard Cite

In this paper, we present the CET-LATS (Compressing Evolution of TINs from Location Aware Time Series) system, which enables testing the impacts of various compression approaches on evolving Triangulated Irregular Networks (TINs). Specifically, we consider the settings in which values measured in distinct locations and at different time instants, are represented as time series of the corresponding measurements, generating a sequence of TINs. Different compression techniques applied to location-specific time series may have different impacts on the representation of the global evolution of TINs-depending on the distance functions used to evaluate the distortion. CET-LATS users can view and analyze compression vs. (im)precision trade-offs over multiple compression methods and distance functions, and decide which method works best for their application. We also provide an option to investigate the impact of the choice of a compression method on the quality of prediction. Our prototype is a web-based system using Flask, a lightweight Python framework, relying on Apache Spark for data management and JSON files to communicate with the front-end, enabling extensibility in terms of adding new data sources as well as compression techniques, distance functions and prediction methods. CCS CONCEPTS • Information systems → Spatial-temporal systems; Web applications.

show abstract

Section: Introductionmentioning

confidence: 99%