Database Reference
In-Depth Information
Figure 14.4.
Sensing in home environments
etc.), and smart displays for information dissemination. These sensors
are embedded in different parts of the home and workplace environment
including on doors, beds, mirrors, bathrooms, mailboxes, in appliances
such as microwaves and allow determining a comprehensive picture of
user activities.
There are several tradeoffs that need to be considered when deciding
how many
smart environment
sensors are needed and where they should
be placed in order to provide enough information for the analysis to ac-
curately recognize activities. While a greater density of sensors provides
more accurate information on the person position and their interactions
with the environment, this comes with increased energy consumption,
cost constraints, and intrusiveness. In addition, increasing sensors lead
to increasing complexity, thus requiring a greater amount of data, large-
scale algorithms, and systems to accurately learn activity models.
Reality mining [72] is also an emerging field complementing ubiqui-
tous healthcare and leveraging data mining technologies. Reality mining
processes all digital information available in the daily environments in
which we evolve these days. Many of the daily activities we perform,
such as checking our email, making phone calls, making a purchase, com-
muting etc., leave digital traces and can be mined to capture records of
our daily experiences. These human physical and social activity traces
are captured by the multitude of sensors in mobile phones, cars, security
cameras, RFID (smart card) readers, road-transport sensors etc. Reality
mining [72], is an emerging field of research that uses statistical anal-
ysis and machine learning methods on these digital traces to develop
comprehensive pictures of our lives, both individually and collectively.
Computational models based on this data, combined with any physio-
