*Go to projects*

Last update: **20/05/2018**

Authors: Manuel Parra-Royon (UGR [SP]), Kiyana Bahadori (U. Pisa [IT]), Ghislain Atemezing (Mondeca [FR]) ,J.M. Benítez (UGR [SP])

Laboratory: http://dicits.ugr.es

Department of: Soft Computing and Intelligent Information Systems , a University of Granada research group

*This website contains a set of documents with additional information about the article not yet published: " Time series workflow modelling with LinkedData".*

This is the documentation for the set of **ontologies** that allow the definition of Time Series (TS) workflow with LinkedData. The name of the scheme proposed is: *tsm-schema* (**T**ime**S**eries **M**odelling **S**chema). The set of schemes allows to define all the key aspects for TimeSeries modelling, such as:

*Times and Dates**Time Series Classes**Univariate modelling**Frecuency analysis**Decomposition and Filtering**Seasonality**Stationary and Co-Integration**N-linear TS analysis**Dynamic Regression**Multivariate Modelling**Clustering and Classification**Continuous time modelling**Functional TS*

Each vocabulary and outline has been defined separately so that the definition of each of the service components, as well as all the additional auxiliary diagrams created, can be worked on more easily.

For each of the schemes, it is available:

- The ontology, defined with TTL, RDF, N3, etc.
- Documentation in W3C format.
- Various datasets of instances of the definitions.

A time series is a collection of observations of a variable taken in a sequential and time-ordered manner (equi-spaced time instants). The series may be annual, half-yearly, quarterly, monthly, etc., depending on the periods of time in which the data it comprises are collected. Time series can be defined as a particular case of stochastic processes, since a stochastic process is a sequence of random, ordered and equidistant variables chronologically referred to a characteristic observable at different times.

Some examples of time series come from fields such as the economy (annual gross domestic product, inflation rate, unemployment rate, etc.), demography (annual births, dependency rate, etc.), meteorology (maximum, average or minimum temperatures, daily rainfall, etc.), etc.

This is the main scheme for time series modeling that allows the definition of time series workflows. This scheme is complemented with *tsmalgo-schema*, which contains all the additional classes that integrate the most commonly used algorithms in time series processing and prediction. Main relationships, classes and entities are defined in the general schema *tsm-schema*

The Turtle (`.ttl`

) format scheme is available in: http://cookingbigdata.com/linkeddata/tsm-schema

Documentation is available here:

http://cookingbigdata.com/linkeddata/tsm-schema/documentation/

Not available yet.

The vocabulary has been submitted to in the LOV platform http://lov.okfn.org/ (LinkedOpenData Vocabularies)

This scheme defines a huge set of algorithms (TimeSeriesAlgorithms-schema), functions and methods (classes) that give meaning to the concepts described in **tsm-schema** and that are necessary for the definition of workflow with Time Series.

The Turtle (`.ttl`

) format scheme is available in: http://cookingbigdata.com/linkeddata/tsmalgo-schema

Documentation is available here:

http://cookingbigdata.com/linkeddata/tsmalgo-schema/documentation/

Not available yet.

The vocabulary has been submitted to in the LOV platform http://lov.okfn.org/ (LinkedOpenData Vocabularies)