Semiqualitative Temporal Patterns in Time-Series Databases

Abstract

A way to obtain behaviour patterns of semiqualitative models of dynamic systems automatically is proposed in this paper. The temporal evolution of these models is stored into a database. This is a time series database. This database may be obtained by means of sensor data or by means of semiqualitative simulations. In any way, the database contains the values of state variables and parameters. Searching for similar patterns in such database is essential, because it helps in predictions hypothesis testing and, in general, in data mining and rule discovery. A language to carry out queries about the qualitative and temporal properties of this time-series database is proposed. The language is also intended to classify the different qualitative behaviours od a model. This classificaction may be carried out according with a specific criterion or automatically by means of clustering algorithms. The semiqualitative behaviour of a system is expressed by means of hierarchical rules obtained by means of machine learning algorithms. The methodology is apllied to a logistics growth model with a delay.