Moving objects beyond raw and semantic trajectories Maria Luisa Damiani Ralf Hartmut Güting University of Milan, I FernUniversität Hagen, D damiani@di.unimi.it rhg@fernuni-hagen.de Fabio Valdés Hamza Issa FernUniversität Hagen, D University of Milan, I fabio.valdes@fernuni- issa@di.unimi.it hagen.de ABSTRACT Somewhat surprisingly, one aspect that is largely ignored Mobile applications, for example for road traffic monitoring, by the most recent literature regards the data management mobile health and animal data ecology, call for methods dimension of semantic trajectories. Put simply: how can we enabling rich and expressive representation of moving ob- store and access semantic trajectories? How can we repre- jects. This demand motivates the increasing concern for the sent semantic trajectories through a rigorous data model? paradigm of semantic trajectories. In this paper, I overview How can semantic trajectories interplay with raw trajec- related research, focusing in particular on the novel data tories and conventional data? These questions have been model of symbolic trajectories proposed for the efficient and only marginally addressed. In fact no operational system flexible handling of semantics-aware trajectories through a enabling the management of semantic trajectories in real Moving Object DBMS. applications exists. We believe that this is a critical limita- tion especially in the light of the increasing availability of big raw trajectory data collected from mobile application- s (e.g. LBS) that creates challenging opportunities for the 1. INTRODUCTION application of this concept. Semantic trajectories is a relatively recent paradigm de- The research that we have undertaken in the context of veloped to provide applications with knowledge about the the European initiative Cost Action MOVE1 aims to fill movement of moving entities. The key idea is to supple- this gap. Indeed the goal is not simply to take some exist- ment the raw mobility data (i.e. raw trajectories in the ing definition of semantic trajectory and find the best way following) - typically sequences of GPS points - with con- for implementing it on a DBMS, but rather to re-think of textual data [4]. For example, semantic trajectories can be the notion of semantically meaningful movement while tar- used to describe the sequence of points of interest visited by geting the specification of a general, formal and operational tourists in a city, or the sequence of transportation means framework. We imagine that in the long run this research used by an individual to reach the working place from home. could lead to the development of a novel class of software Basically a semantic trajectory consists of a raw trajectory platforms for mobility data handling. The users of these augmented with annotations regarding the whole trajecto- systems will be able to organize and analyze mobility da- ry or parts of it. Probably because of its simplicity and ta in the same way that users now organize and analyze naturalness, the concept of semantic trajectory has attract- spatial data in a conventional GIS platform, e.g. Quantum ed the interest of numerous researchers over the last years. GIS, or using one of the more recent platforms on cloud, Current research develops along diverse streams including: e.g. GISCloud. While the idea in itself may sound not par- ontology/conceptual modeling, mobility pattern mining for ticularly innovative, just a restyling of GIS, we believe that the generation of semantic annotations, semantic location these platforms, going beyond the notion of Moving Object privacy, and - more recently - the connection with the the- DBMS, can greatly facilitate the development of novel and ories of complex networks and social analysis. The main challenging applications. In what follows, the notion of se- results achieved so far are nicely summarized in the survey mantic trajectory is presented; next the concept of symbolic paper [4]. trajectory is introduced along with the results achieved so far and major open issues. 2. SEMANTIC TRAJECTORIES Early work on semantic trajectories was triggered by the experimental analysis of a set of raw trajectories about a group of birds [5]. By using the standard functionalities of a GIS, we found that the sequences of points, just pairs 1 http://move-cost.info/ Proceedings IMMoA’13 41 http://www.dbis.rwth-aachen.de/IMMoA2013/ of timestamped coordinates, associated with birds identi- We have defined a simple generic data model able to cap- fiers were actually representing the migration routes from ture different types of semantics called symbolic trajectory Central Europe to Africa and vice versa. Such discovery, [6]. In essence the idea is to represent semantic informa- that was somewhat unexpected, inspired the proposal of a tion in terms of names or labels. For example an activity novel model for the high level representation of movemen- (running, walking ) and points of interest (Colosseum, Lou- t. Since this first result, research developed along different vre) can be straightforwardly described by labels while sen- directions, including the following: sor readings, e.g. temperature, need first to be turned into qualitative values, e.g. high, low temperature. Formally, a • Conceptual modeling. The first conceptualization was symbolic trajectory is an ordered sequence of pairs centered on the notions of stop and move [5]. A stop represents a temporary suspension of the movement, (i1 l1 ), ..(in ln ) while a move is the transfer from one stop to anoth- called units when each unit uj = (ij lj ) consists of a time in- er stop. While this conceptualization is appropriate terval ij and a label lj . The label lj describes the movement in many applications, there is increasing evidence that in the time interval ij . Symbolic trajectories are provided stop-and-move is just one of the possible mobility pat- as abstract data types and integrated into the ADT model terns. For example Yan et al. [7] present an approach defined in [3]. For example a symbolic trajectory describ- to extract and represent the sequence of activities from ing places and the transportation means used to reach those raw trajectories. In the light of these experiences, a places, can be as follows: novel conceptual model has been recently proposed which enables the attachment of any kind of mean- (2013-01-17-9:02:30 2013-01-17-9:05:51) "home") ing (not just stop and move) to sequences of points (2013-01-17-9:05:51 2013-01-17-9:08:44) "bus") [4]. (2013-01-17-9:08:44 2013-01-17-9:50:02) "train") (2013-01-17-9:50:02 2013-01-17-17:50:02) "work") • Extraction of mobility patterns. A major research di- .... rection regards the mining of mobility patterns to au- tomatically annotate semantic trajectories. Early work The core technical contribution is a novel language for pat- by Alvares et al. [1] focuses on the identification of tern matching and rewriting on symbolic trajectories. The stops and moves. Numerous approaches can be found pattern language enables the extraction of subsequences from in literature, either explicitly related to the notion of symbolic trajectories. Patterns are defined as regular ex- stop-and-move or developed within different commu- pressions that can be matched by single units or sequences nities. A comprehensive survey can be found in [4]. of units. For example, the query: Which are the trajecto- ries in which the individuals take more than 1 hour to move • The privacy of mobility patterns. A different issue is from home to work? can be solved specifying the following to preserve the privacy of sensitive mobility patterns pattern: such as the presence in places, e.g. hospitals and reli- *(_ home ) Z* (_ work)*// getDuration(Z.time)> 3600 gious buildings, that might reveal sensitive information about moving individuals. This problem is particular- where: ly challenging in on-line applications, e.g. LBS and geo-social networks, whereas the privacy mechanism - Z is a variable denoting a sequence of units, the symbol has to rely on partial knowledge of the movement (past * denotes a sequence of zero or more units, and current positions are known, but not future posi- - ( home)Z ∗ ( work) is the pattern tions). The privacy of mobility patterns in an open issue [2]. An approach in this direction, focused on - getDuration(X.time) > 3600 is the condition that the protection of specific mobility pattern, i.e. sensi- must be met by the matching sequences, in this case tive places, is presented in [8]. the duration in seconds of the transfer from home to work. 3. SYMBOLIC TRAJECTORIES An important feature of the language is that it is embed- Semantic trajectories are often considered the result of an ded into an existing Moving Object DBMS (i.e. Secondo). analytical process conducted on raw trajectories. We believe The pattern language at work is illustrated in a video2 . that the notion of semantic trajectory is valuable on its own, independently of how these trajectories are generated. For 4. CONCLUDING REMARKS example, annotations can be deliberately attached by indi- viduals (e.g. user can specify the transportation means) or Capturing and representing the meaning of movement is a even the annotation can be automatically attached by the challenging issue that calls for novel solutions. We are work- location tracking system (e.g. locations in indoor settings ing on the definition of the symbolic trajectory data model have natural semantics, such as room 1 and building A). for the representation of time-varying textual descriptions. Moreover, even in those cases in which semantic trajecto- A number of issues are still open. For example, a major is- ries are obtained from an analytical process, the problem sue is integrating - whenever it is meaningful - the symbolic remains of how to encode them in a machine readable for- dimension with the geometric dimension of the movemen- m. This is the focus of our current research that we briefly t. Another major issue regards the usability of the system present in what follows. that is fundamental for an effective deployment of symbolic trajectories in real applications. 3.1 The data model 2 http://molle.fernuni-hagen.de/DfnA/SymbolicTrajectories.mp4 Proceedings IMMoA’13 52 http://www.dbis.rwth-aachen.de/IMMoA2013/ 5. REFERENCES modeling and analysis. ACM Comput. Surv., [1] L. Alvares, V. Bogorny, B. Kuijpers, B. de Macedo, 45(4):42:1–42:32, Aug. 2013. J.and Moelans, and A. Vaisman. A model for enriching [5] S. Spaccapietra, C. Parent, M.L. Damiani, trajectories with semantic geographical information. In J. de Macedo, F. Porto, and C. Vangenot. A conceptual Proc. ACM GIS, GIS ’07, 2007. view on trajectories. Data Knowl. Eng., 65(1):126–146, [2] M.L. Damiani. European Data Protection : Coming of 2008. Age ?, chapter Privacy enhancing techniques for the [6] F. Valdés, M.L. Damiani, and R. Güting. Symbolic protection of mobility patterns in LBS: research issues trajectories in secondo: Pattern matching and and trends. Springer, 2013. rewriting. In DASFAA, 2013. [3] R. H. Güting, M. Böhlen, M. Erwig, C. Jensen, [7] Z. Yan, D. Chakraborty, C. Parent, S. Spaccapietra, N. Lorentzos, M. Schneider, and M.Vazirgiannis. A and K. Aberer. Semitri: a framework for semantic foundation for representing and querying moving annotation of heterogeneous trajectories. In Proc. objects. ACM Trans. Database Syst., 25(1):1–42, 2000. EDBT, 2011. [4] C. Parent, S. Spaccapietra, C. Renso, G. Andrienko, [8] E. Yigitoglu, M.L. Damiani, O. Abul, and C. Silvestri. N. Andrienko, V. Bogorny, M.L. Damiani, Privacy-preserving sharing of sensitive semantic A. Gkoulalas-Divanis, J. Macedo, N. Pelekis, locations under road-network constraints. In IEEE Y. Theodoridis, and Z. Yan. Semantic trajectories MDM, 2012. Proceedings IMMoA’13 63 http://www.dbis.rwth-aachen.de/IMMoA2013/