A Semantic Model for Movement Data Warehouses

Abstract

Despite recent progresses in methods for processing data about the movement of objects in the geographic space, some fundamental issues remain unresolved. One of them is how to describe movement segments (e.g., semantic trajec- tories, episodes like stops and moves) and diverse movement patterns (e.g., moving clusters, hotel-restaurant-shop-hotel), with formal semantic descriptions. Another issue is how to arrange descriptive data and measures in a Movement Data Warehouse (MDW) for powerful information analyses and reasonable performance. This paper introduces general def- initions for movement segments, movement patterns, their categories and hierarchies. The proposed constructs are se- mantically enriched with references to concepts (categories) and/or instances of these concepts (objects) arranged in dis- tinct hierarchies. Based on these constructs, we propose a semantic multidimensional model for MDW. A case study illustrates the expressiveness of the proposal for analyzing movement data collected via social media and semantically enriched with Linked Open Data (LOD)

The Baquara2 knowledge-based framework for semantic enrichment and analysis of movement data

“…The movement data structuring in movement segment hierarchies and the construction of movement analysis 337 facets of KB resources (concepts or objects) and their semantic relationships can take place before, during, or after this semantic en-338 richment process. These issues are beyond the scope of this article, and addressed in other works[34,19].…”

“…Therefore, to realize potential applications it is necessary to develop appropriate 33 methods to semantically enrich movement data. The recent progresses in movement data handling [7][8][9][10][11] include plenty of signifi- 34 cant contributions for structuring and analyzing movement data, but are mainly based on just spatio-temporal data. Notwithstanding, 35 it is recognized by the scientific community that semantic issues, including the exploitation of textual and contextual information that 36 can come associated with spatio-temporal coordinates, must be addressed yet to better understand and exploit movement data…”

“…(TBox) and an extensional level (ABox) based on description logics [33]. The first has at least one conceptualization hierarchy (classes 242 organized according to their IS_A relationships), and the latter at least one object hierarchy (instances that can be organized by some analysis in data warehouses can be found in [34]. Besides the Baquara pre-defined facets described in the following, that work also 245 proposes facets based on movement movement segment hierarchies, movement pattern hierarchies, and subsumption hierarchies 246 of their respective categories.…”

“…MovementMonitoringDevice (e.g., CellPhone), as discussed in Section 2. facets can be found in [34]. That work exploits facets based on movement movement segment hierarchies, movement pattern hier-278 archies, and subsumption hierarchies of their respective categories to support simpler and more efficient query expressions than 279 the ones referring just to facets like the ones described in this paper.…”

Enriching and Analyzing Semantic Trajectories With Linked Open Data

“…( However, we have recently witnessed a growing research area where the representation of movement is by the so-called semantic trajectories (FILETO et al, 2014;BOGORNY et al, 2014;PARENT et al, 2013). A semantic trajectory is a representation of a trajectory as a sequence of meaningful segments.…”