=Paper=
{{Paper
|id=Vol-2646/17-paper
|storemode=property
|title=MASTER: A Multiple Aspects View on Trajectories
|pdfUrl=https://ceur-ws.org/Vol-2646/17-paper.pdf
|volume=Vol-2646
|authors=Ronaldo dos Santos Mello,Vania Bogorny,Luis Otavio Alvares,Luiz Henrique Zambom Santana,Carlos Andres Ferrero,Angelo Augusto Frozza,Geomar Andre Schreiner,Chiara Renso
|dblpUrl=https://dblp.org/rec/conf/sebd/MelloBASFFSR20
}}
==MASTER: A Multiple Aspects View on Trajectories==
https://ceur-ws.org/Vol-2646/17-paper.pdf
MASTER: A Multiple Aspects View on
Trajectories
Discussion Paper
Ronaldo dos Santos Mello1 , Vania Bogorny1 , Luis Otavio Alvares1 ,
Luiz Henrique Zambom Santana1 , Carlos Andres Ferrero1,2 ,
Angelo Augusto Frozza1,3 , Geomar Andre Schreiner1 , and Chiara Renso4
1
Universidade Federal de Santa Catarina - PPGCC, Florianopolis, Brazil
2
Instituto Federal de Santa Catarina, Lages-SC, Brazil
3
Instituto Federal Catarinense, Camboriu-SC, Brazil
4
ISTI-CNR, Pisa, Italy
Abstract. For many years trajectory data have been treated as se-
quences of space-time points or stops and moves. However, with the
explosion of the Internet of Things (IoT) and the flood of Big Data gen-
erated on the Internet, like weather channels and social network inter-
actions, which can be used to enrich mobility data, trajectories become
more and more complex, with multiple and heterogeneous data dimen-
sions that can be integrated with trajectories. In this paper we introduce
multiple aspect trajectories and we propose a robust conceptual and
logical data model and a storage solution for efficient multiple aspect
trajectory queries. The main strength of our data model is the com-
bination of simplicity and expressive power to represent heterogeneous
aspects, ranging from simple labels to complex objects. We evaluate the
proposed model in a tourism scenario.
1 Introduction
With the explosion of the Internet of Things (IoT) and the flood of Big Data
generated on the Internet, like weather channels and social network interactions
(e.g., Flickr, Facebook, Twitter, Foursquare), it is now possible to collect huge
volumes of movement data about people, animals, and objects like cars, buses,
drones, etc. Sensors installed either indoor (e.g., smart homes) or outdoor al-
low the collection of data about the place, like temperature, air pollution, noise,
luminosity, etc, or about the object that is moving around or inside this place,
like the heart rate (with a smart watch), the emotional status (with a micro-
phone that analyses the voice intonation), blood pressure, sleeping stages, etc.
By collecting all these information we have a new type of movement data, i.e.,
a trajectory enriched with different semantic aspects.
Copyright c 2020 for this paper by its authors. Use permitted under Creative Com-
mons License Attribution 4.0 International (CC BY 4.0). This volume is published
and copyrighted by its editors. SEBD 2020, June 21-24, 2020, Villasimius, Italy.
Fig. 1. An example of a multiple aspect trajectory
We can observe from Figure 1 that a trajectory became a complex object with
numerous data dimensions that are contextual to the movement and heteroge-
neous in the form, which we define in this paper as aspects. The more aspects
we have, the more complete is the representation of the real movement of an ob-
ject, and more useful and interesting information we can infer about objects and
places. The challenge is how to integrate all these heterogeneous information in
a single trajectory representation, and the main questions we want to answer in
this paper are: (i) Is it possible to define a data model that is simple in structure,
but generic enough to represent any aspect related to the movement, and covers a
large number of applications? (ii) Is there a way to efficiently query and extract
patterns from data represented in this model?
We claim that multiple aspects represent a new view over trajectories, and
a new paradigm concerning mobility data. These aspects are not only simple
semantic labels, but may be complex objects and/or heterogeneous information
intrinsically associated to the physical traces of the moving objects.
In this discussion paper we highlight the results of a published paper [4].
In that paper we introduced the concept of multiple aspect trajectory and pro-
pose a novel approach for modeling this kind of trajectories called MASTER.
MASTER comprises a conceptual and a logical data model for multiple aspect
trajectories, as well as a storage solution that is very appropriate for multiple as-
pect trajectory queries. In the full paper we also compared the MASTER model
with a competitor DB called SECONDO[5]. Due to the lack of space, we omit
here this comparison and we refer the reader to the full paper for details. The
main novelties of MASTER compared to other state of the art works (detailed
described in the full paper [4]) is that MASTER is more generic since we intro-
duce the notion of ”aspect” and how these aspects can enrich the location data.
Other approaches do not support relationships between moving objects and do
not propose a solutions for storing and querying huge volumes of trajectories
and aspects or moving objects.
2 The MASTER Model
The main strength of our conceptual model is the combination of simplicity and
expressive power for representing aspects. An aspect may be related to a moving
object, to the entire trajectory or any trajectory point, and may hold any type
of data, ranging from simple labels to complex objects.
For the logical model, we consider a graph-based representation (the RDF
standard [6]) that is generic enough to model trajectories and aspects extracted
from heterogeneous data sources, like geolocated structured record files and ge-
olocated social media posts (e.g., tweets). Finally,
we consider NoSQL databases for efficient storage and retrieval of large
amounts of trajectory data. Our inspiration comes from the polyglot persistence
approach [7], which states that a conceptual data model can be split and mapped
to several database models for maximizing query performance.
We introduce a conceptual data model for multiple aspect trajectories, which
is shown in Figure 2. We start the description of the model with the new concept
of aspect.
An aspect is a real world fact that is relevant for trajectory data analysis,
and it is characterized by an aspect type. For instance, the aspect train belongs
to an aspect type transportation mode, and an aspect rainy belongs to an aspect
type weather condition. An aspect type has a set of attributes and it may also
be a subtype of a more general aspect type, allowing the modeling of an aspect
type subtypeOf hierarchy, like POI ←accommodation←hotel. More formally, an
aspect type is defined as follows.
Definition 1. An Aspect Type asptype = (desc, ATT, aspsupertype ) is a catego-
rization of a real-world fact with a description desc, a set of attributes ATT =
{a1 , a2 , . . . , az } that hold its properties, and a (possibly empty) supertype aspect
aspsupertype .
An aspect type and its attributes act as a metadata definition for an aspect.
As a consequence, an aspect is always related to at least one aspect type and
its attributes. For example, given an aspect type weather condition, some of its
attributes could be temperature, wind speed and climate. In the following, we
define an aspect.
Definition 2. An Aspect asp = (desc, SAT) is a relevant real-world fact, where
desc is the aspect description, and SAT = {at1 , at2 , ..., atx } is a non-empty set
of aspect types, with their attributes and respective values, which the aspect may
hold, being ati = (asptype k , AT Vk ), ati ∈ SAT , a tuple with an aspect type
asptype k and a non-empty set AT Vk = {a1 : v1 , a2 : v2 , . . . , an : vn } of attribute-
value pairs so that each pair (ai : vi ) ∈ AT Vk is an instantiation of a property
ai of asptype k with a (atomic or multivalued) value vi .
An aspect definition supports numbers, ranges, text, geometries (when an
aspect describes, for example, the shape of a hurricane at a specific time instant),
or any type of complex object.
Definition 3. A Semantic Meaning SM = (asp, asptype ) is an association be-
tween an aspect asp and an aspect type asptype that gives the context of the
aspect, so that asptype belongs to the aspect types of the aspect asp.
Fig. 2. The Conceptual Model for Multiple Aspect Trajectories
An aspect with a semantic meaning can be associated to a multiple aspect
trajectory, a trajectory point, a moving object, or a relationship between moving
objects in our conceptual model (see Figure 2). When an aspect varies frequently
during the object movement, the aspect with its semantic meaning is associated
to each trajectory point and it is called volatile aspect (VA). Some examples are
the visited places (or stops) and the heart rate. An aspect is also associated to
a point when it represents a sparse and instant happening, like a social media
post or check-in. When an aspect does not change during an entire trajectory,
it is called a long term aspect (LTA) and is associated to the multiple aspect
trajectory. Examples of this kind of aspect are the town on which the trajectory
occurs or the person occupation. When an aspect holds during the entire life of
an object, it is called a permanent aspect (PA) and is associated to the object
and not to the trajectory. One example is the person birthplace. These aspect
categories are directly related to the query performance. Queries on volatile
aspects, i.e., queries related to trajectory points, will be more time consuming,
while long term and permanent aspects will be retrieved more quickly.
Based on these foundations, we now define a multiple aspect trajectory.
Definition 4. A Multiple Aspect Trajectory mat = (P, S LTA, mo, desc) is
a sequence of points P = hp1 , p2 , . . . , pn i of a moving object mo, a (possible
empty) set of long term aspects S LTA, being S LTA = {sm1 , sm2 , ..., smp } a
set of semantic meanings, and a description desc, with pi = (xi , yi , ti , S VA),
pi ∈ P, being x and y the spatial position of mo at the time instant t, and S VA
the set of volatile aspects related to pi , where S VA = {sm1 , sm2 , ..., smq } is a
set of (possible empty) semantic meanings.
A multiple aspect trajectory belongs to a moving object. A moving object is
any entity that moves along space and time. This object is always associated
to a type, which can be a person, a drone, an animal, a car, or even a natural
phenomenon, like a hurricane. We formally define it in the following.
Definition 5. A Moving Object mo = (motype , desc, S PA) is an entity that
can physically move in space and time, having a description desc, a set of (pos-
sible empty) permanent aspects S PA, being S PA = {sm1 , sm2 , ..., smr } a set
of semantic meanings, and a type motype that categorizes it.
A new feature in MASTER when compared to the state-of-the-art data mod-
els for trajectories is the moving object relationship. A moving object may hold
any type of relationship with other objects, and these relationships may also be
characterized by different aspects such as the type of relationship (e.g., friend-
ship, professional, family). We define a moving object relationship in Definition
6.
Definition 6. A Moving Object Relationship mor = (mo1 , mo2 , S RA) is a
relevant association between two moving objects mo1 and mo2 that holds a (pos-
sible empty) set of relationship aspects S RA, being S RA = {sm1 , sm2 , ..., sms }
a set of semantic meanings.
Finally, we model spatial features and events. The first one denotes any rele-
vant POI that is not spatially related to trajectory points, so it is not an aspect.
Instead, it means any POI located in the trajectory neighborhood, like a nearby
restaurant. In Figure 1, an example of spatial feature is the church located be-
tween the POIs work and restaurant. So, when a trajectory point intersects a
relevant POI, it is modeled as an aspect. Otherwise, it is a spatial feature. Spatial
features are useful for answering spatial queries like which are the restaurants
located at a distance less than α from the trajectory of object A? or which trajec-
tories have an envelope whose area is higher than avenue B? Similarly, an event
denotes a happening that does not have a relationship with trajectories, but it
is relevant for queries that investigate events in the trajectory neighborhood. An
event occurs at a spatial feature and is valid for the period that it happened.
For defining the MASTER logical model we adopt the Resource Descrip-
tion Framework (RDF) [9] as our logical data model because RDF data can
be modeled as a graph, which is a flexible data structure to represent the high
heterogeneity of possible aspects, as well as the great number of aspect relation-
ships with trajectories, points and moving objects. Besides, on using RDF we
are consonant with the Semantic Web standards of WWW Consortium (W3C)
for publishing and manipulating data on the Web [10].
Figure 3 shows the proposed logical model, where dotted arrows represent an
entity-attribute relationship, continuous arrows represent relationships between
entities, and the ellipsis represent entities or attributes. A predicate label fol-
lowed by a cardinality pair denotes a multivalued relationship. One example is
a point that may be enriched with zero to several semantic meanings. An RDF
triple schema in such a modeling is represented by two ellipsis connected by an
arrow. One example is a moving object (subject) that is the owner (predicate)
of a multiple aspect trajectory (object).
The conversion of the conceptual model to a logical schema in RDF was
inspired by several related approaches [2, 1, 3], which propose the following map-
ping rules:
Fig. 3. The Logical Model for Multiple Aspect Trajectories
– an entity is converted to a node;
– an attribute of an entity (or relationship) erm is converted to a node nt , and
an edge is defined from erm to nt in order to connect them;
– a relationship between entities is converted to an edge that connects the
entities, or an intermediate node between the entities.
There is only one rule for the mapping of entities and attributes, so their
conversion is straightforward. Even entities without attributes, like MovingOb-
jectRelationship (see Figure 2), became nodes in the logical model because they
have relationships with other entities.
We decided to consider the conversion to an edge if the relationship has no
properties related to it, and the conversion to a node otherwise. On doing so,
we avoid the generation of too many nodes in the RDF schema, and only the
relationships that hold semantic meaning and hasValue as properties (see Figure
2) were mapped to nodes, and the second one was renamed to Value for sake of
understanding. We also decide to maintain only the connections of Aspect and
Attribute with the Value node to avoid a redundant edge between Aspect and
Attribute.
The adopted storage solution for maintaining data represented in the MAS-
TER logical model. This solution is called Rendezvous [8]. Rendezvous is a
triplestore based on NoSQL databases for querying large RDF datasets. NoSQL
databases have been proposed for managing big data efficiently [7]. Therefore,
as multiple aspect trajectories are highly heterogeneous and multidimensional
data, NoSQL databases are a suitable storage resource for this new type of
trajectory. Compared to related work, Rendezvous was chosen due to its multi-
We define a label hasValue for this predicate to identify it as an entity-attribute or
relationship-attribute connection.
model NoSQL support for storing RDF data and its efficient processing of typical
SPARQL queries. Rendezvous manages RDF data in a distributed database.
3 Discussion and future works
The MASTER model has been evaluated over two perspectives: (i) a qualitative
analysis at the conceptual level by modeling a tourism application, as well as an
evaluation at the logical level to attest that an RDF-based storage strategy is
suitable to answer the main types of multiple aspect trajectory queries; and (ii)
a quantitative evaluation at the storage level by comparing the query running
time performance of our storage solution with a baseline. With this experiment
we show the feasibility of MASTER as a data model that can be efficiently stored
and accessed. All details are presented in the full paper [4].
We claim that three main types of queries can be posed to multiple aspect
trajectories: (i) queries that return moving objects (e.g., which are the moving
objects that were born in Florianopolis and are male? ); (ii) queries that return
trajectories (e.g., which are the trajectories that stayed at an accommodation
place? ); and (iii) queries that return aspects (e.g., which accommodations were
visited in Paris by persons that were born in Florianopolis and are male? ). These
three queries are examples of star-shaped, chain-shaped and complex queries,
respectively.
Table 1 shows these queries written in SPARQL or GeoSPARQL with an ar-
bitrary complexity depending on the number of entities that must be considered
to generate the query result, which allows an efficient processing by our RDF
storage solution.
Future works include a performance evaluation over larger data sets of en-
riched trajectories, as well as the evaluation of other Big Data storage technolo-
gies, such as NewSQL databases, for maintaining multiple aspect trajectories.
We also intend to extend MASTER to model data analytics information over
multiple aspect trajectories, considering, for instance, dependencies among as-
pects.
Although out of the scope of this paper, it is also very important to consider
privacy issues that these kinds of enriched trajectories might pose. When com-
bining the different semantic aspects to the location information, privacy breach
might happen. It is therefore crucial to develop privacy preserving methods to
guarantee privacy when multiple aspects are involved.
Acknowledgments
This work was supported by the Brazilian agencies CAPES and CNPq, and the
MASTER Project which received funding from the European Union’s Horizon
2020 research and innovation programme under the Marie Sklowdoska-Curie
agreement N. 777695.
Table 1. Typical SPARQL Queries over Multiple Aspect Trajectories
SELECT mo? SELECT t? SELECT a?
WHERE { WHERE { WHERE {
mo? is-a ’Person’ t? composedOf p? a? rdfs:type ’Aspect’
mo? enrichedWith sm? p? enrichedWith sm? a? is-a at?
sm? aspect ’Florianopolis’ sm? aspectType at? at? subtypeOf ’Accommodation’
sm? aspectType ’birthplace’ at? subtypeOf sm? aspectType a?
mo? enrichedWith sm2? ’Accommodation’} p? enrichedWith sm?
sm2? aspect ’Male’ FILTER(
sm2? aspectType ’Gender’} geof:sfIntersects(?p, Paris))
t? composedOf p?
mo? owner t?
mo? is-a ’Person’
mo? enrichedWith sm?
sm? aspect ’Florianopolis’
sm? aspectType ’birthplace’
mo? enrichedWith sm2?
sm2? aspect ’Male’
sm2? aspectType ’Gender’}
(a) Star (b) Chain (c) Complex
References
1. S. Bagui and J. Bouressa. Mapping rdf and rdf-schema to the entity relation-
ship model. Journal of Emerging Trends in Computing and Information Sciences,
5(12):953–961, 2014.
2. M. Choi, C. Moon, and D. Baik. Transformation of a relational database to
RDF/RDFS with er2 idm. IEICE Transactions, 96-D(7):1478–1488, 2013.
3. G. Daniel, G. Sunyé, and J. Cabot. Umltographdb: Mapping conceptual schemas
to graph databases. In Conceptual Modeling - 35th International Conference, ER
2016, Gifu, Japan, November 14-17, 2016, Proceedings, pages 430–444, 2016.
4. R. dos Santos Mello, V. Bogorny, L. O. Alvares, L. H. Z. Santana, C. A. Ferrero,
A. A. Frozza, G. A. Schreiner, and C. Renso. MASTER: A Multiple Aspect View
on Trajectories. Trans. GIS, 23(4):805–822, 2019.
5. R. H. Güting, V. T. de Almeida, D. Ansorge, T. Behr, Z. Ding, T. Höse, F. Hoff-
mann, M. Spiekermann, and U. Telle. SECONDO: an extensible DBMS platform
for research prototyping and teaching. In 21st ICDE 2005, pages 1115–1116, 2005.
6. J. Z. Pan. Resource description framework. In Handbook on Ontologies, pages
71–90. Springer, 2009.
7. P. J. Sadalage and M. Fowler. NoSQL distilled : a brief guide to the emerging world
of polyglot persistence. Addison-Wesley, 2013.
8. L. H. Z. Santana and R. dos Santos Mello. Workload-aware rdf partitioning and
sparql query caching for massive rdf graphs stored in nosql databases. In Brazilian
Symposium on Databases (SBBD), 2017, pages 184–195. SBC, 2017.
9. W3C. RDF - Semantic Web Standards, 2018.
10. W3C. Semantic Web - W3C, 2018.