A data mining approach for fall detection by using k-nearest neighbour algorithm on wireless sensor network data
Küçük Resim Yok
Tarih
2012
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
INST ENGINEERING TECHNOLOGY-IET
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Fall detection technology is critical for the elderly people. In order to avoid the need of full time care giving service, the actual trend is to encourage elderly to stay living autonomously in their homes as long as possible. Reliable fall detection methods can enhance life safety of the elderly and boost their confidence by immediately alerting fall cases to caregivers. This study presents an algorithm of fall detection, which detects fall events by using data-mining approach. The authors' proposed method performs detection in two steps. First, it collects the wireless sensor network (WSN) data in stream format from sensor devices. Second, it uses k-nearest neighbour algorithm, that is, well-known lazy learning algorithm to detect fall occurrences. It detects falls by identifying the fall patterns in the data stream. Experiments show that the proposed method has promising results on WSN data stream in detecting falls.
Açıklama
Anahtar Kelimeler
Kaynak
IET COMMUNICATIONS
WoS Q Değeri
Q3
Scopus Q Değeri
Q2
Cilt
6
Sayı
18
