Build user daily activity model and model structure changing

Abstract

Abstract-because of the activity dynamic it is a challenged work to build daily activity model of user. The goal of the paper is to build daily activity model of the user with high accuracy. At first the raw data derived from motion detector will be "translated" to state data, then use state split and merge to build the basic model. Thirdly in order to increase accuracy of the model the count of the state data increased by changing the parameter of the translator. Here we get another two activity models with higher accuracy.The changing of the count of the state data caused to the structure changing of the build activity model. We merge the states from same route of activity models, compare different models and find out the same activity trends of the user. On the other hand from the models with higher accuracy the hidden activity will be detected.

Activity recognition using a hierarchical model

“…This feature helps in deciding if the person is moving in the house. Doing so much work to extract only a single feature might look like to much work to do however, the created topology is later utilized when extracting dist ← 1/similarity 12: if dist < minDist then 13: minDist ← dist (V 1 , V 2 ) ← nearest vertices that are not already connected 6:…”

“…, 3, 2). So, the time of day features take on values in range [0][1][2][3][4][5][6][7][8][9][10][11][12] and are of length 24.…”

“…Smart environments that are capable of activity recognition [2]- [6] and assistive technologies for the disabled [7], [8] has been designed. In the scope of ATTEND (AdapTive scenario recogniTion for Emergency and Need Detection) project, we are working on a system that is capable of activity recognition using a network of sensors installed in the living environment.…”