Fall detection by using K-nearest neighbor algorithm on WSN data
2010 IEEE Globecom Workshops, GC 2010, Florida, Amerika Birleşik Devletleri, 5 - 10 Aralık 2010, ss.2054-2058, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/glocomw.2010.5700306
- Basıldığı Şehir: Florida
- Basıldığı Ülke: Amerika Birleşik Devletleri
- Sayfa Sayıları: ss.2054-2058
- Anahtar Kelimeler: Data mining, Fall detection, KNN algorithm, Wireless sensor networks
- Maltepe Üniversitesi Adresli: Evet
Özet
Falls are serious problem especially for elderly people. Day by day the elderly people are living alone and the children of these people want to get information in dangerous situations. With the alarm systems, someone in difficulty can be detected and emergency aid can be sent. We propose a system to detect falls by using a data mining approach on WSN data. The proposed system evaluated using data stream collected from sensor device and fall detection accuracy and precision are calculated. Our solution demonstrated promising results on WSN data stream. ©2010 IEEE.