A Comparative Analysis of the Impact of Features on Human Activity Recognition with Smartphone Sensors

Sousa, Wesllen; Souto, Eduardo; Rodrigres, Jonatas; Sadarc, Pedro; Jalali, Roozbeh; El-Khatib, Khalil

doi:10.1145/3126858.3126859

Cited by 32 publications

(28 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On the other hand, [13] submitted a combination between feature selection techniques and a deep learning method, concretely a Deep Belief Network (DBN), with some good results, higher than the ones achieved with SVM-based models, which showed to be one of the best algorithms to use in HAR problematics. By contrast, in [14,15] they made comparisons between different feature selections for different widely used machine learning (ML) algorithms in the literature. Results showed that frequency-based features are more feasible, at least for algorithms like SVM or CNN, as they throw the best results.…”

Section: Related Workmentioning

confidence: 99%

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

García-González

Rivero

Fernández-Blanco

et al. 2020

Sensors

109

View full text Add to dashboard Cite

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.

show abstract

Section: Related Workmentioning

confidence: 99%

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

García-González

Rivero

Fernández-Blanco

et al. 2020

Sensors

109

View full text Add to dashboard Cite

show abstract

“…These results are not new; Sousa et al [ 9 ], for example, reached the same conclusion and they affirm that time domain features are sufficient to represent activities from inertial sensors’ data, especially using algorithms of the decision tree family where they have a lower processing cost compared to the KNN and SVM.…”

Section: Experiments and Resultsmentioning

confidence: 69%

“…To simplify the presentation of the results, the experiments were performed only with the inertial sensor accelerometer data. According to Sousa et al [ 9 ] this sensor is sufficient to represent the physical activities of users. In the others, the configurations used specifically for baselines are based on results of experiments carried out in the literature [ 3 , 9 , 46 ], as: (1) time window size: 5 s; (2) overlap: 50%; (3) features: time domain, frequency domain derived from FFT, and Wavelet.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…This advancement has benefited applications such as health care [ 1 , 2 ], monitoring physical activities [ 3 , 4 ], domestic activities [ 5 ], and safety [ 6 , 7 ]. As a result, many solutions have been proposed based on machine learning algorithms, especially the shallow algorithms (e.g., SVM, Decision Tree, Naive Bayes and KNN) [ 4 , 8 , 9 ] and, more recently, the deep learning algorithms based on neural networks (e.g., CNN, RNN, RBM, SAE, DFN and DBM) [ 10 , 11 , 12 ]. The main difference between the two approaches lies in the way the data is prepared.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Human Activity Recognition Based on Symbolic Representation Algorithms for Inertial Sensors

Lima

Bragança

Quispe

et al. 2018

Sensors

View full text Add to dashboard Cite

Mobile sensing has allowed the emergence of a variety of solutions related to the monitoring and recognition of human activities (HAR). Such solutions have been implemented in smartphones for the purpose of better understanding human behavior. However, such solutions still suffer from the limitations of the computing resources found on smartphones. In this sense, the HAR area has focused on the development of solutions of low computational cost. In general, the strategies used in the solutions are based on shallow and deep learning algorithms. The problem is that not all of these strategies are feasible for implementation in smartphones due to the high computational cost required, mainly, by the steps of data preparation and the training of classification models. In this context, this article evaluates a new set of alternative strategies based on Symbolic Aggregate Approximation (SAX) and Symbolic Fourier Approximation (SFA) algorithms with the purpose of developing solutions with low computational cost in terms of memory and processing. In addition, this article also evaluates some classification algorithms adapted to manipulate symbolic data, such as SAX-VSM, BOSS, BOSS-VS and WEASEL. Experiments were performed on the UCI-HAR, SHOAIB and WISDM databases commonly used in the literature to validate HAR solutions based on smartphones. The results show that the symbolic representation algorithms are faster in the feature extraction phase, on average, by 84.81%, and reduce the consumption of memory space, on average, by 94.48%, and they have accuracy rates equivalent to conventional algorithms.

show abstract

“…Globally the smart phone user count is estimated to reach 2.87 billion by the year 2020 [1]. Modern smart phones are more powerful in terms of processing, communication and data acquisition from built-in sensors [2], thus making it an integral part of pervasive computing. Applications such as health care assistance, health monitoring, M-commerce, home automation, smart environments, human machine interaction and so on are focused on the effective utilization of smart phone sensor data.…”

Section: Introductionmentioning

confidence: 99%

Classification of Activities of Daily living using Smart Phones

Ranjith*¹,

Pillai²

2019

IJRTE

View full text Add to dashboard Cite

Abstract: Smart phones have become an integral part of everyday human life. These phones are packed with various sensors for different purposes. Most of them are used for understanding the environment in which the user uses the phone so that the device could respond rapidly. Indirectly the phone extracts context information of the users like the activity performed using accelerometer and gyroscope sensors. This information can be used for a variety of applications like home automation, smart environment, etc to perform automatic changes to the environment without direct input from the user. This paper deals with the classification of activities of daily living like walking, jogging, sitting, standing, upstairs and downstairs using the data collected from accelerometer sensor within the smart phone. A comparative analysis has been performed on different machine learning techniques for activity classification.

show abstract

A Comparative Analysis of the Impact of Features on Human Activity Recognition with Smartphone Sensors

Cited by 32 publications

References 17 publications

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

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

Human Activity Recognition Based on Symbolic Representation Algorithms for Inertial Sensors

Classification of Activities of Daily living using Smart Phones

Contact Info

Product

Resources

About