2011
DOI: 10.1016/j.neucom.2011.06.023
|View full text |Cite
|
Sign up to set email alerts
|

An efficient feature selection method for mobile devices with application to activity recognition

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0
6

Year Published

2012
2012
2023
2023

Publication Types

Select...
3
3
3

Relationship

1
8

Authors

Journals

citations
Cited by 21 publications
(22 citation statements)
references
References 51 publications
0
16
0
6
Order By: Relevance
“…Par exemple, Peng et coll. [15] présentent un algorithme fonctionnant sur un appareil mobile qui permet d'extraire certaines caractéristiques utiles à la détection d'activités. Dans le même ordre d'idée, Fuentes et coll.…”
Section: La Detection Des Dangersunclassified
“…Par exemple, Peng et coll. [15] présentent un algorithme fonctionnant sur un appareil mobile qui permet d'extraire certaines caractéristiques utiles à la détection d'activités. Dans le même ordre d'idée, Fuentes et coll.…”
Section: La Detection Des Dangersunclassified
“…Automatic recognition of physical activity was previously investigated using Support Vector Machine (SVM) classification [31]- [34] and Fast Artificial Neural Network (FANN) classification [35]. In particular, fall detection was also investigated using only the accelerometer in a cell phone [36]- [38].…”
Section: B Softwarementioning
confidence: 99%
“…For data classification a two-stage feature selection algorithm has been developed [6]. A fast two-stage subset selection algorithm is combined with a model-based variable selection approach to give an efficient algorithm applicable for smartphones.…”
Section: Algorithmic Techniquementioning
confidence: 99%
“…However Peng et al [6] have developed a new approach to automatic activity classification, which is practically memory efficient and capable of real-time classification on a smartphone without the need to calculate the algorithm on an external device. Other applications that employ an activity recognition algorithm such as CenceMe [7] for social networking, the MSP and its real-world deployments as described in [8] or sensor networking systems such as those in [9], either require processing on back-end servers, are external custommade devices, or require extra external sensors than those provided within the phone itself.…”
Section: Introductionmentioning
confidence: 99%