Implementation of a Semantic Language for Environmental Phenomena Processing in Time Series

Abstract

This paper describes our implementation of a semantic language for environmental phenomena processing in time series. Our contribution consists of the dedicated time series processing language Formula3, which can be used in a dynamic environment by expert users and time series analysts, and is based on a novel pipe-oriented architecture. In addition to the language description, we also present our approaches for key-value-pair time series processing and semantic enrichment of time series expressions. The concepts are validated by the implementation of Forumla3 as a processing and query language and the time series are visualized in the graphical time series editor. We conclude the paper with references to related work and an outlook to future extensions for our language.