Title :
A Dynamic Key Context Information Based Scenes Recognition Method
Author :
Hui-juan Zhang ; Xin-jie Chen
Author_Institution :
Sch. of Software Eng., Tongji Univ., Shanghai, China
Abstract :
In the context-awareness problems, classification models are used to recognize the user scenes based on the sensor signal data. To improve the classification accuracy, a Dynamic Key Context Information Based scenes recognition method is proposed in this paper. Since the location information can be easily obtained by the sensors, the data set is divided into groups by their location information. An attribute selection technique is applied in each group to identify the key information.
Keywords :
pattern classification; sensors; ubiquitous computing; attribute selection technique; classification models; context-awareness problems; dynamic key context information based scene recognition method; location information; scene recognition; sensor signal data; user scene recognition; Accuracy; Artificial neural networks; Context; Context modeling; Legged locomotion; Niobium; Training; Scene Clusters; context-awareness; key context information;
Conference_Titel :
Wireless Communication and Sensor Network (WCSN), 2014 International Conference on
Print_ISBN :
978-1-4799-7090-2
DOI :
10.1109/WCSN.2014.69
