|
Title:
|
THE ACTIVITY RECOGNITION: ADVANTAGEOUS BUT CHALLENGING! |
|
Author(s):
|
Chaima Bouali, Abderrahim Tahiri and Olivier Habert |
|
ISBN:
|
978-989-8533-77-7 |
|
Editors:
|
Mário Macedo and Piet Kommers |
|
Year:
|
2018 |
|
Edition:
|
Single |
|
Keywords:
|
Activity Recognition, Smart Homes, Healthcare |
|
Type:
|
Reflection Paper |
|
First Page:
|
277 |
|
Last Page:
|
280 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The Activity Recognition (AR) has caught the attention of a number of researchers in the last years. This topic has been highlighted due to its usability in different applications such as healthcare, security systems, etc. The three main categories of the approaches depicted in different papers are: Supervised, unsupervised and semi-supervised methods. However, several challenges are faced while developing a trustworthy system for the identification of activities. In this document, a presentation of different approaches related to the AR field will take place and also a discussion about the different issues that lead to an accuracy reduction of the methods developed in the past 10 years. The aim of our study has been the extraction of the lacks and the pros from each method in order to build a reliable AR system that takes into consideration the different challenges. |
|
|
|
|
|
|
