<?xml version="1.0" encoding="UTF-8"?>
<XML><RECORDS>
<RECORD>
	<REFERENCE_TYPE>3</REFERENCE_TYPE>
	<AUTHORS>
		<AUTHOR>Miyuki Shiraishi</AUTHOR>
		<AUTHOR>Yasuyuki Washio</AUTHOR>
		<AUTHOR>Chihiro Takayama</AUTHOR>
		<AUTHOR>Vili Lehdonvirta</AUTHOR>
		<AUTHOR>Hiroaki Kimura</AUTHOR>
		<AUTHOR>Tatsuo Nakajima</AUTHOR>
	</AUTHORS>
	<YEAR>2009</YEAR>
	<TITLE>Tracking behavior in persuasive apps: is sensor-based detection always better than user self-reporting?</TITLE>
	<SECONDARY_TITLE>CHI EA '09: Proceedings of the 27th international conference extended abstracts on Human factors in computing systems</SECONDARY_TITLE>
	<PLACE_PUBLISHED>New York, NY, USA</PLACE_PUBLISHED>
	<PUBLISHER>ACM</PUBLISHER>
	<PAGES>4045–4050</PAGES>
	<ISBN>978-1-60558-247-4</ISBN>
	<ABSTRACT>This paper aims to discuss the roles for the two types of tracking user behavior. Considering these two types of tracking, sensor based recognition has a great advantage when sensing human activity, but it is not always adequate when tracking in the real world. In this paper, we compare the benefits and drawbacks of sensor-based tracking versus self-reported data in persuasive applications called EcoIsland.</ABSTRACT>
</RECORD>
</RECORDS></XML>