Presentation 2012/5/10
Improving Fault Tolerance of Wearable Sensor-based Activity Recognition Technique by Completing Defective Feature Value
RYOMA UCHIDA, REN OMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Activity recognition techniques with wearable sensors are one of the most important technologies to achieve context-aware systems. However, existing recognition techniques are lack of fault tolerance in case of sensors data loss, such as communication disconnection and sensor failure. Our study proposes two methods to improve the fault tolerance by completing missing feature values with avairable some sensor. Our methods use multiple regression and the kernel method. We had some experiments to verify the usefulness of proposed methods. As a result, our method is confirmed to be able to improve fault tolerance of activity recognition technique. F-Measure rate is 69.8% in case of sensors data loss. Our method's F-Measure rate of multiple regression and kernel method are 70.7% and 71.1% that improved more than 1.0% than case of sensor data loss.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Activity recognition / improving fault tolerance / completing defective value
Paper # Vol.2012-UBI-34 No.31
Date of Issue

Conference Information
Committee USN
Conference Date 2012/5/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Ubiquitous and Sensor Networks(USN)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Improving Fault Tolerance of Wearable Sensor-based Activity Recognition Technique by Completing Defective Feature Value
Sub Title (in English)
Keyword(1) Activity recognition
Keyword(2) improving fault tolerance
Keyword(3) completing defective value
1st Author's Name RYOMA UCHIDA
1st Author's Affiliation Toyohashi University of Technology()
2nd Author's Name REN OMURA
2nd Author's Affiliation Toyohashi University of Technology
Date 2012/5/10
Paper # Vol.2012-UBI-34 No.31
Volume (vol) vol.112
Number (no) 31
Page pp.pp.-
#Pages 8
Date of Issue