A Smart Home Application for Resident Activity Prediction

Abstract

Predicting resident activity in smart home has become an emerging research trend in the field of Pervasive computing. Resident activity prediction by using contextual data in smart home provides a resident intended service thereby makes him more flexible. In this paper, an effort has been made to achieve resident activity prediction by developing an android based smart home application which takes the following contextual parameters: resident id, location & status of the devices as input and predicts the activity of the resident as the output. The proposed smart home application involves the following steps: 1) Constructing the training dataset with the contextual parameters such as resident id, location & status of the device 2) Storing the captured data in SQL database 3) Retrieving the activity from the database by using Python programming. This can be achieved by using the Bluetooth module that communicates between the Raspberry Pi & mobile.