A Public Domain Dataset for Real-Life Human Activity Recognition Using Smartphone Sensors

Abstract

In recent years, human activity recognition has become a hot topic inside the scientific community. The reason to be under the spotlight is its direct application in multiple domains, like healthcare or fitness. Additionally, the current worldwide use of smartphones makes it particularly easy to get this kind of data from people in a non-intrusive and cheaper way, without the need for other wearables. In this paper, we introduce our orientation-independent, placement-independent and subject-independent human activity recognition dataset. The information in this dataset is the measurements from the accelerometer, gyroscope, magnetometer, and GPS of the smartphone. Additionally, each measure is associated with one of the four possible registered activities: inactive, active, walking and driving. This work also proposes asupport vector machine (SVM) model to perform some preliminary experiments on the dataset. Considering that this dataset was taken from smartphones in their actual use, unlike other datasets, the development of a good model on such data is an open problem and a challenge for researchers. By doing so, we would be able to close the gap between the model and a real-life application.

iSPLInception: An Inception-ResNet Deep Learning Architecture for Human Activity Recognition

“…We use a variation of the UCI HAR smartphone dataset [18], in our work. This work has been extended by D. Garcia-Gonzalez et al [21] with a new dataset that only contains four activities of active, inactive, walking, and driving and data from the accelerometer, gyroscope, magnetometer, and global positioning system (GPS) of the smartphone. Thu and Han [22] address the presence of postural transitions within another variation of this dataset and propose a way to deal with them when trying to improve the classification of the activities.…”

The Use of Inertial Measurement Units for the Study of Free Living Environment Activity Assessment: A Literature Review

“…It now includes features used in the biomedical and research fields (geolocation, magnetometer, accelerometer, high quality camera, etc.). It is a widely used devices, used both professionally [ 86 , 114 ] and personally. The smartphone is already used through the development of applications for medical purposes.…”

“…This disparity may be a hindrance to research, which justifies the need for unification. One way to unify the studies would be the creation of a public and universally accessible dataset, as in the recent study by Garcia et al [ 86 ]. Additional ways to unify the studies could be to work on a common goal such as the challenges (see, for example, in [ 121 ]), to have a shared aim for the whole community [ 121 ].…”

“…Of all studies using more than one sensor, 67% use four or less sensors. [ 27 , 28 , 54 , 61 , 65 , 74 , 86 ]). Moreover, only three studies use more than ten IMUs—these studies use the IGS-180 suit consisting of seventeen IMUs [ 18 , 63 , 70 ].…”

“…It should be noted that some studies tried to detect groups of activities (sedentary activities [ 38 , 62 , 75 ], stationary activities [ 78 , 86 , 93 ], etc.) which included several specific activities (standing, sitting, lying down for the group of so-called stationary activities, etc.).…”