VTT TECHNICAL RESEARCH CENTRE OF FINLAND
www.vtt.fi


                                             ContextCapture
     Using Context-based Awareness Cues to Create Story-like Status Updates
                                Ville Antila (ville.antila@vtt.fi), Jussi Polet (jussi.polet@vtt.fi)




The prototype consists of a mobile application and a
web application, which is integrated with Facebook
and Twitter. The mobile application gathers context
data from the device itself, available sensors and
from nearby devices via Bluetooth. The sensed
context information is presented to the user along
with proposals for other descriptions, which have
been used to describe similar contexts earlier. After
selecting the context items and their abstractions, the
status update is sent to the server application, which
stores the received context data into a semantic
model and creates a story-like status update for
Facebook and / or Twitter.


Goals

 The application demonstrates technical aspects
 of collaborative context e.g. how the contextual
 information can be exchanged between different
 devices and used for UX-related testing issues.
 Machine learning can be enabled by saving the
 selected abstractions and associated raw data.
 This will create a collectively created database of
 context semantics that can be shared between
 users. The stored data can be used for future
 purposes, such as the basis for situation-aware
 recommendations.


Context gathering                                                   Solution

 The context recognition is based on different                        • mobile application and a related web applica-
 sensors of activity such as :                                          tion integrated with Facebook and Twitter
 • the accelerometer                                                  • easy and fast addition of context-awareness
 • ambient light sensor                                                 cues into status updates
 • GPS data                                                           • pre- and user-defined context descriptions
 • open applications on the device                                      and five most used description recommenda-
 • device system information                                            tions
 • nearby Bluetooth devices
 • WLAN access points                                                 Cross-platform smartphone support :
                                                                      • Android 2.2 +
 Context descriptions are shown to the user based                     • MeeGo
 on this data.                                                        • Symbian^3 (also Symbian S60 5th)

UbiComp2011: ContextCapture (Poster)

  • 1.
    VTT TECHNICAL RESEARCHCENTRE OF FINLAND www.vtt.fi ContextCapture Using Context-based Awareness Cues to Create Story-like Status Updates Ville Antila (ville.antila@vtt.fi), Jussi Polet (jussi.polet@vtt.fi) The prototype consists of a mobile application and a web application, which is integrated with Facebook and Twitter. The mobile application gathers context data from the device itself, available sensors and from nearby devices via Bluetooth. The sensed context information is presented to the user along with proposals for other descriptions, which have been used to describe similar contexts earlier. After selecting the context items and their abstractions, the status update is sent to the server application, which stores the received context data into a semantic model and creates a story-like status update for Facebook and / or Twitter. Goals The application demonstrates technical aspects of collaborative context e.g. how the contextual information can be exchanged between different devices and used for UX-related testing issues. Machine learning can be enabled by saving the selected abstractions and associated raw data. This will create a collectively created database of context semantics that can be shared between users. The stored data can be used for future purposes, such as the basis for situation-aware recommendations. Context gathering Solution The context recognition is based on different • mobile application and a related web applica- sensors of activity such as : tion integrated with Facebook and Twitter • the accelerometer • easy and fast addition of context-awareness • ambient light sensor cues into status updates • GPS data • pre- and user-defined context descriptions • open applications on the device and five most used description recommenda- • device system information tions • nearby Bluetooth devices • WLAN access points Cross-platform smartphone support : • Android 2.2 + Context descriptions are shown to the user based • MeeGo on this data. • Symbian^3 (also Symbian S60 5th)