=Paper=
{{Paper
|id=None
|storemode=property
|title=Short Paper: Annotating Microblog Posts with Sensor Data for Emergency Reporting Applications
|pdfUrl=https://ceur-ws.org/Vol-839/crowley.pdf
|volume=Vol-839
|dblpUrl=https://dblp.org/rec/conf/semweb/CrowleyPB11
}}
==Short Paper: Annotating Microblog Posts with Sensor Data for Emergency Reporting Applications==
https://ceur-ws.org/Vol-839/crowley.pdf
Short Paper: Annotating Microblog Posts with
Sensor Data for Emergency Reporting
Applications
David N. Crowley1,2 , Alexandre Passant2 , and John G. Breslin1,2
1
School of Engineering and Informatics
National University of Ireland, Galway
firstname.lastname@nuigalway.ie
2
Digital Enterprise Research Institute
National University of Ireland, Galway
firstname.lastname@deri.org
Abstract. The explosion in user-generated content (on the Social Web)
published from mobile devices has seen microblog platforms like Twitter
grow exponentially. Twitter is a microblogging platform founded in 2006,
which by October 2010 had roughly 175m users and as of June 2011,
Twitter processed 200m posts per day. Twitter data has been utilised to
predict/report natural disasters, civil unrest, and media topics. Smart-
phones and other mobile devices contain an array of sensors but are
under-utilised on the Social Web. In this paper, we propose a method
for annotating microblog posts with multi-sensor data by representing it
with ontologies such as SSN and SIOC. We present an alignment of these
ontologies and outline an enhanced Twitter client that would allow users
to enter an emergency mode where all or most of the available sensor
data would be published as annotations to the users post, allowing relief
organisations to use any data relevant.
Keywords: SSN, Microblog, SIOC, Citizen, Sensors, Social Sensing
1 Introduction
The unprecedented 96% growth in smartphone sales1 and in user numbers on
social platforms like Twitter (572,000 new accounts created on March 12, 2011)
demonstrate the growth in the use of the mobile web. As microblogging lends
itself to instantaneous updates, data related to events occurring around the world
is created before it can be reported on by more traditional media methods or
even by blog or blog-like services. In parallel with this growth in mobile-based
microblogging, mobile devices themselves have begun to incorporate increasing
amounts of sensors for various purposes, ranging from detecting light levels when
a phone is placed close to one’s head to accelerometers that can detect orientation
changes and movement in various directions.
1
http://www.gartner.com/it/page.jsp?id=1466313
In this short paper, we look at using microblogging platforms as citizen sens-
ing/reporting platforms by adding mobile sensor data to user posts and de-
scribing that data using the SSN (Semantic Sensor Network)2 ontology and the
SIOC (Semantically-Interlinked Online Communities)3 ontology. In particular,
we outline applications for emergency scenarios, where people can report on
events using microblogging while automatically attaching all available sensor
data from their mobile devices (in order to provide context to emergency re-
ports). The structure of this paper is as follows. Section 2 will describe related
work in this area along with a brief review of mobile sensors. We will describe
the Twitter Annotations initiative in Section 3, and how it can be used for
sensor data annotations. Section 4 will detail the alignments required between
the social and sensor data ontologies SIOC and SSN. Section 5 will outline our
proposed ’emergency mode’ microblogging client that allows users to upload all
available sensor data with a post to aid relief workers/government agencies. We
will present conclusions in Section 6.
2 Related Work
Sheth uses the example of Twitter posts during the Mumbai terrorist attacks in
November 2008 when Twitter updates and Flickr feeds by citizens using mobile
devices reported observations of these events in real time[1].4 Twitter data has
been used in event/disaster reporting and prediction[2]. Tapia et al. examined the
usage of Twitter to aid relief workers with information regarding disasters, and
they saw one method of using “microblogged data as ambient or contextual data
to enrich the information provided to the NGO at the time of disaster”[3]. Mobile
devices contain many sensor formats that provide information such as location
(through GPS or cell tower locations) to create/add context to microblog posts
and status updates. Companies like Foursquare use this contextual data to create
various geo-social gaming/marketing applications.
In relation to microblog posts, at present GPS adds location to the data of
the post made, but in the field of multi-sensor context awareness, researchers
are currently examining ways to augment devices with an awareness of their
situation and environment to add contextual understanding through the use of
combined sensor data. As Gellersen et al. asserts “Position is a static environment
and does not capture dynamic aspects of a situation”[4], and this concept can
be applied to most single sensor data, but with multi-sensor context awareness
the diverse sensor readings are combined and then with processing situational
awareness can be derived. Situation awareness is the observation of surround-
ing elements/events in relation to the user, this perception of the immediate
environment lets humans derive meaning and aids in decision making.5
2
http://www.w3.org/2005/Incubator/ssn/XGR-ssn/
3
http://sioc-project.org/
4
http://www.telegraph.co.uk/news/worldnews/asia/india/3530640/
Mumbai-attacks-Twitter-and-Flickr-used-to-break-news-Bombay-India.html
5
http://en.wikipedia.org/wiki/Situation_awareness
Table 1. Mobile Sensing Types
Sensor Types Sensor Return Values
Accelerometer Acceleration along X, Y, Z axes (m/s2 )
Gyroscope Angular speed along X, Y, Z axes measured in radians/second
Magnetic Field Magnetic field in X, Y, Z axes measured in micro-Tesla µT
Orientation Angle measurement along X, Y, Z axes in degrees
Proximity Distance (cm)
NFC A short-range wireless technology
GPS Returns location if available (longitude and latitude)
Camera Captures still images and video
Microphone Allows for capture of audio
Compass Standard directional compass values
Light Light intensity in Lumens
Table 1 shows a non-exhaustive list of common mobile device sensors and
their expected return values. Apple, Google, Nokia, and Microsoft have devel-
oped APIs in their mobile operating systems to allow access to these sensors,
which allow developers to use sensor data in their applications. Sensor APIs al-
low application developers access to sensor readings to aid in user experience but
also to allow for the collection of context data [5]. The structure for attaching
sensor readings to microblog posts can take the format of Twitter annotations,
which we will describe in the next section, or SSN-annotated SIOC posts for
semantically-enhanced applications as we will describe later on.
3 Twitter Annotations
Twitter Annotations is an initiative from Twitter that allows additional struc-
tured metadata to be attached to tweets, going beyond the geotemporal annota-
tions normally found in social media content. While the annotation or metadata
is structured, it is open to the user or developer to decide what additional infor-
mation is attached to the microblog post. There is an overall limit of 512 bytes
for the metadata payload, but this may be expanded as usage increases.
As an example in JSON, data about a movie described in a tweet could be
attached to the tweet using the annotation {“movie”:{“title”:“The Guard”}},
indicating that the title of the movie is “The Guard”. The guidelines for Twit-
ter Annotations state that the goal is to “bring more structured data to tweets
to allow for better discovery of data and richer interactions.”6 In the sphere
of citizen sensing, Twitter annotations can be seen as a way to standardise an
emerging field of supplementing microblog posts with sensor data and, as with
any area, standardisation is important. Figure 1 illustrates two examples of an-
notations in the Twitter Annotations JSON format. The first example describes
a digital compass sensor in an Android mobile device that returns direction in
degrees, and the second describes data returned from a three-axis accelerometer.
6
http://dev.twitter.com/pages/annotations_overview
In this work, the Twitter Annotations format will be used for adding sensor/
multi-sensor data to tweets using the Twitter Annotations API, and will inspire
how we attach sensor data (represented using the SSN ontology) to tweets, blog
posts and other microblog posts described via SIOC.
Fig. 1. Annotation Examples
4 Aligning the SIOC and SSN ontologies
SIOC allows the semantic interlinking of content items from forums, blogs and
other social websites, and aims to enable the integration of online community
information[6]. SIOC provides a Semantic Web ontology for representing rich
data from the Social Web using the Resource Description Framework (RDF).
By describing the social data contained within online communities (powered by
blogs, wikis, and forums) using semantic technologies, SIOC enables this data
to become a “Social Web of Data”[7].
SNN
Microblog ObservationValue
has_sensor_data
has_container hasValue some
MicroblogPost SensorOutput
isProducedBy some
Sensor
Fig. 2. Aligning the SSN Ontology with the SIOC Ontology
SIOC was originally written to describe web-based discussion on blogs and
message boards, but with the SIOC Types module this has been expanded to in-
clude items like Microblog and MicroblogPost. SIOC has received significant
adoption in commercial and open-source software applications7 : it has been
adopted in the core of Drupal 7 and around 100 applications use SIOC.
Figure 2 outlines our method for annotating microblog posts with
sensor/multi-sensor data by representing it with ontologies such as SSN
and SIOC. The property has sensor data will join sioct:MicroblogPost to
ssn:ObervationValue. We proprose to create a SIOC Sensors (siocs) module to in-
clude this and future related properties. The sioct:MicroblogPost itself can have
one or more ObservationValue(s). Figure 3 is an example of a microblog post
with orientation sensor data attached. We define an AndroidOrientation sensor
that has a defined SensorOutput that has value OriObservationValue a subclass
of ObservationValue and has three properties hasXQuantityValue, hasYQuan-
tityValue, and hasZQuantityValue, defined in a Citizen Sensors ontology (cs).
Fig. 3. RDF Example: Android Orientation Sensor Annotation
5 Scenario
We will now describe a scenario whereby data from multiple sensors can be at-
tached to microblog posts using the aforementioned alignments to aid in emer-
gency scenarios. We are currently developing a semantic microblogging client
for the Android platform that implements both Twitter Annotations and SSN-
annotated SIOC posts for emergency reporting with sensor data.
In an emergency, the user could employ the semantic microblogging client
and activate the emergency mode that would allow the application to annotate
any available sensor data to their post (including photos). The available sensor
readings could help emergency workers by attaching the direction the user is
facing, noise levels in the surrounding area, light levels, direction of movement,
and any other available data to the post. If GPS is unavailable, then from these
sensors and the information extracted from the microblog post (place names or
points of interest) an estimated location along a directional line could be calcu-
lated. In situations where a snapshot of data is not relevant, attaching aggregated
7
http://sioc-project.org/applications
values/lists of values describing changes in activity, compass direction, and noise
levels over time might better communicate the user’s situation. Furthermore, the
microblog post contents and the annotated sensor readings could aid emergency
teams with reports including direction and lighting conditions.
6 Conclusion
By combining the Social Web and sensors, applications can provide an extension
of social activities through sensors, as user activity is modelled by both voluntary
user input and sensor data annotated to the posts. In this paper, we describe how
this will be implemented using web standards like the SIOC onotology and by
aligning SIOC with the SSN ontology to both describe users’ posts semantically
and attach contextual sensor data to the post through metadata annotations.
We have described a scenario that uses this combined SIOC-SSN representation,
based on a semantic microblogging client currently being developed for mobile
devices that will enable emergency reporting functionality.
7 Acknowledgements
This work has been supported in part by Science Foundation Ireland under grant
number SFI/08/CE/I1380 (Lı́on 2). We would like to thank Fabrizio Orlandi and
Myriam Leggieri for their input.
References
1. Sheth, A.: Citizen Sensing, Social Signals, and Enriching Human Experience. Inter-
net Computing, IEEE 13(4), 87–92 (2009)
2. Sakaki, T., Okazaki, M., Matsuo, Y.: Earthquake Shakes Twitter Users: Real-Time
Event Detection by Social Sensors, Proceedings of the 19th International Conference
on World Wide Web, pp. 851–860. Raleigh, North Carolina, USA (2010)
3. Tapia, A.H., Bajpai, K., Jansen, B.J., Yen, J.: Seeking the Trustworthy Tweet: Can
Microblogged Data Fit the Information Needs of Disaster Response and Humanitar-
ian Relief Organizations, Proceedings of the 8th International ISCRAM Conference,.
pp. 1-10, Lisbon, Portugal (2011)
4. Gellersen, H.W., Schmidt, A., Beigl, M.: Multi-Sensor Context-Awareness in Mo-
bile Devices and Smart Artefacts. Mobile Networks and Applications 7(5), 341–351
(2002)
5. Lane, N.D., Miluzzo, E., Hong Lu, Peebles, D., Choudhury, T., Campbell, A.T.: A
Survey of Mobile Phone Sensing. Mobile Networks and Applications 48(9), 140–150
(2010)
6. Berrueta, D., Brickley, D., Decker, S., Fernández S., Görn, C., Harth, A., Heath,
T. , Idehen, T., Kjernsmo, K., Miles, A., Passant, A., Polleres, A., Polo, L., Sintek,
M.: SIOC Core Ontology Specification. W3C Member Submission 12 June 2007,
http://www.w3.org/Submission/sioc-spec/
7. Bojars, U., Passant, A., Cyganiak, R., Breslin, J.G.: Weaving SIOC into the Web
of Linked Data, Proceedings of the WWW 2008 Workshop on Linked Data on the
Web (LDOW2008), Beijing, China (2008)