Human Activity Recognition on Mobile Devices Using Artificial Hydrocarbon Networks

Ponce, Hiram; Miralles-Pechuán, Luis; González Mora, José Guillermo; Martinez-Villaseñor, Lourdes

doi:10.1007/978-3-030-02837-4_2

Human Activity Recognition on Mobile Devices Using Artificial Hydrocarbon Networks

Journal

Advances in Soft Computing

Lecture Notes in Computer Science

ISSN

0302-9743

1611-3349

Date Issued

2018

Author(s)

Miralles-Pechuán, Luis

González Mora, José Guillermo

Type

Resource Types::text::book::book part

DOI

10.1007/978-3-030-02837-4_2

URL

https://scripta.up.edu.mx/handle/20.500.12552/4294

Abstract

Human activity recognition (HAR) aims to classify and identify activities based on data-driven from different devices, such as sensors or cameras. Particularly, mobile devices have been used for this recognition task. However, versatility of users, location of smartphones, battery, processing and storage limitations, among other issues have been identified. In that sense, this paper presents a human activity recognition system based on artificial hydrocarbon networks. This technique have been proved to be very effective on HAR systems using wearable sensors, so the present work proposes to use this learning method with the information provided by the in-sensors of mobile devices. Preliminary results proved that artificial hydrocarbon networks might be used as an alternative for human activity recognition on mobile devices. In addition, a real dataset created for this work has been published. © Springer Nature Switzerland AG 2018.