Trajectory data warehouses: storing and aggregating frequent ST patterns

In this paper we present an approach for storing and aggregating spatio-temporal patterns by using a Tra jectory Data Warehouse (TDW). In particular, our aim is to allow the analysts to quickly evaluate frequent patterns mined from tra jectories of moving ob jects occurring in a specific spatial zone and during a given temporal interval. We resort to a TDW, based on a data cube model, having spatial and tem- poral dimensions, discretized according to a hierarchy of regular grids, and whose facts are sets of tra jectories which intersect the ST cells of the cube. The idea is to enrich such a TDW with a new measure: frequent patterns obtained from a data-mining process on tra jectories. As a consequence these patterns can be analysed by the user at various levels of granularity with the use of OLAP queries. The research issues discussed in this paper are (1) the extraction/mining of the patterns to be stored in each cell, which requires an adequate pro jection phase of tra jectories before mining; (2) the ST aggregation of patterns to answer roll-up queries, which poses many problems due to the holistic nature of the aggregation function.