Data analyzing and daily activity learning with hidden Markov model

Author

Yin, GuoQing ; Bruckner, Dietmar

Author_Institution

Inst. of Comput. Technol., Vienna Univ. of Technol., Vienna, Austria

Volume

3

fYear

2010

fDate

22-24 Oct. 2010

Abstract

To observe and analyze person´s daily activities, and build the activities model is an important task in an intelligent environment. In an Ambient Assisted Living (AAL) project we get sensor data from a motion detector. At first we translate and reduce the raw data to state data. Secondly using hidden Markov model, forward algorithm, and Viterbi Algorithm to analyze the data and build the person´s daily activity model. Comparing individual observation with the build model to find out best and worst (abnormal) activities.

Keywords

data analysis; hidden Markov models; image motion analysis; object detection; sensor fusion; unsupervised learning; Viterbi algorithm; ambient assisted living project; daily activity learning; data analysis; forward algorithm; hidden Markov model; intelligent environment; motion detector; person daily activity model; sensor data; unsupervised learning; Acoustics; Buildings; Hidden Markov models; Intelligent environment; Viterbi algorithm; forward algorithm; hidden Markov model (HMM);

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Application and System Modeling (ICCASM), 2010 International Conference on

Conference_Location

Taiyuan

Print_ISBN

978-1-4244-7235-2

Electronic_ISBN

978-1-4244-7237-6

Type

conf

DOI

10.1109/ICCASM.2010.5620212

Filename

5620212