|
Title:
|
A SELF LEARNING CONTEXT-AWARE DOMOTICS SYSTEM TO AUTOMATE USER ACTIONS |
|
Author(s):
|
Niels Pardons, Natalie Kcomt Ché, Yves Vanrompay, Yolande Berbers |
|
ISBN:
|
978-972-8939-30-4 |
|
Editors:
|
Hans Weghorn, Pedro Isaías and Radu Vasiu |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
Domotics, ambient intelligence, embedded, user action automation |
|
Type:
|
Full Paper |
|
First Page:
|
96 |
|
Last Page:
|
102 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Home automation or domotics systems are ambient intelligence systems that are designed to help people proactively, but sensibly. In this paper we propose a system that learns and automates patterns in the interactions of the user with the home automation devices. We show our approach and architecture. An event processing tool is used to handle the events from the home automation devices, prediction algorithms predict the next action and both rule-based algorithms as reinforcement learning decide which actions are suitable to be automated. We show the results of our system on both a synthetic data set and a real data set. The automation system manages to automate a significant number of interactions for the user. |
|
|
|
|
|
|
