Modular Development of Ontologies for Provenance in Detrending Time Series

The scientific knowledge, in many areas, is obtained from time series analysis, which is usually done in two phases, preprocessing and data analysis. Trend extraction (detrending) is one important step of preprocessing phase, where many detrending software using different statistical methods can be applied for the same time series to correct them. In this context, the knowledge about time series data is relevant to the researcher to choose appropriate statistical methods to be used. Also the knowledge about how and how often the time series were corrected is essential for choice of detrending methods that can be applied to getting better results. This knowledge is not always explicit and easy to interpret. Provenance using Web Ontology Language - OWL ontologies contributes for helping the researcher to get knowledge about data and processes executed. Provenance information allows knowing as data were detrended, improving the decision making and contributing for generation of scientific knowledge. The main contribution of this paper is presenting the modular development of ontologies combined with Open Provenance Model - OPM, which is extended to facilitate the understanding about as detrending processes were executed in time series data, enriching semantically the preprocessing phase of time series analysis.