This article describes an investigation to determine the optimal placement of accelerometers for the purpose of detecting a range of everyday activities. The paper investigates the effect of combining data from accelerometers placed at various bodily locations on the accuracy of activity detection. Eight healthy males participated within the study. Data were collected from six wireless tri-axial accelerometers placed at the chest, wrist, lower back, hip, thigh and foot. Activities included walking, running on a motorized treadmill, sitting, lying, standing and walking up and down stairs. The Support Vector Machine provided the most accurate detection of activities of all the machine learning algorithms investigated. Although data from all locations provided similar levels of accuracy, the hip was the best single location to record data for activity detection using a Support Vector Machine, providing small but significantly better accuracy than the other investigated locations. Increasing the number of sensing locations from one to two or more statistically increased the accuracy of classification. There was no significant difference in accuracy when using two or more sensors. It was noted, however, that the difference in activity detection using single or multiple accelerometers may be more pronounced when trying to detect finer grain activities. Future work shall therefore investigate the effects of accelerometer placement on a larger range of these activities.

Data annotation is a time-consuming process posing major limitations to the development of Human Activity Recognition (HAR) systems. The availability of a large amount of labeled data is required for supervised Machine Learning (ML) approaches, especially in the case of online and personalized approaches requiring user specific datasets to be labeled. The availability of such datasets has the potential to help address common problems of smartphone-based HAR, such as inter-person variability. In this work, we present (i) an automatic labeling method facilitating the collection of labeled datasets in free-living conditions using the smartphone, and (ii) we investigate the robustness of common supervised classification approaches under instances of noisy data. We evaluated the results with a dataset consisting of 38 days of manually labeled data collected in free living. The comparison between the manually and the automatically labeled ground truth demonstrated that it was possible to obtain labels automatically with an 80–85% average precision rate. Results obtained also show how a supervised approach trained using automatically generated labels achieved an 84% f-score (using Neural Networks and Random Forests); however, results also demonstrated how the presence of label noise could lower the f-score up to 64–74% depending on the classification approach (Nearest Centroid and Multi-Class Support Vector Machine).

The galvanic skin response (GSR; also widely known as electrodermal activity (EDA)) is a signal for stress-related studies. Given the sparsity of studies related to the GSR and the variety of devices, this study was conducted at the Human Health Activity Laboratory (H2AL) with 17 healthy subjects to determine the variability in the detection of changes in the galvanic skin response among a test group with heterogeneous respondents facing pleasant and unpleasant stimuli, correlating the GSR biosignals measured from different body sites. We experimented with the right and left wrist, left fingers, the inner side of the right foot using Shimmer3GSR and Empatica E4 sensors. The results indicated the most promising homogeneous places for measuring the GSR, namely, the left fingers and right foot. The results also suggested that due to a significantly strong correlation among the inner side of the right foot and the left fingers, as well as the moderate correlations with the right and left wrists, the foot may be a suitable place to homogenously measure a GSR signal in a test group. We also discuss some possible causes of weak and negative correlations from anomalies detected in the raw data possibly related to the sensors or the test group, which may be considered to develop robust emotion detection systems based on GRS biosignals.

Activity recognition (AR) from an applied perspective of ambient assisted living (AAL) and smart homes (SH) has become a subject of great interest. Promising a better quality of life, AR applied in contexts such as health, security, and energy consumption can lead to solutions capable of reaching even the people most in need. This study was strongly motivated because levels of development, deployment, and technology of AR solutions transferred to society and industry are based on software development, but also depend on the hardware devices used. The current paper identifies contributions to hardware uses for activity recognition through a scientific literature review in the Web of Science (WoS) database. This work found four dominant groups of technologies used for AR in SH and AAL—smartphones, wearables, video, and electronic components—and two emerging technologies: Wi-Fi and assistive robots. Many of these technologies overlap across many research works. Through bibliometric networks analysis, the present review identified some gaps and new potential combinations of technologies for advances in this emerging worldwide field and their uses. The review also relates the use of these six technologies in health conditions, health care, emotion recognition, occupancy, mobility, posture recognition, localization, fall detection, and generic activity recognition applications. The above can serve as a road map that allows readers to execute approachable projects and deploy applications in different socioeconomic contexts, and the possibility to establish networks with the community involved in this topic. This analysis shows that the research field in activity recognition accepts that specific goals cannot be achieved using one single hardware technology, but can be using joint solutions, this paper shows how such technology works in this regard.

This article presents a study on analyzing body movements by using a single accelerometer sensor. The investigated categories of body movements belong to the Laban Effort Framework: Strong—Light, Free—Bound and Sudden—Sustained. All body movements were represented by a set of activities used for data collection. The calculated accuracy of detecting the body movements was based on collecting data from a single wireless tri-axial accelerometer sensor. Ten healthy subjects collected data from three body locations (chest, wrist and thigh) simultaneously in order to analyze the locations comparatively. The data was then processed and analyzed using Machine Learning techniques. The wrist placement was found to be the best single location to record data for detecting Strong—Light body movements using the Random Forest classifier. The wrist placement was also the best location for classifying Bound—Free body movements using the SVM classifier. However, the data collected from the chest placement yielded the best results for detecting Sudden—Sustained body movements using the Random Forest classifier. The study shows that the choice of the accelerometer placement should depend on the targeted type of movement. In addition, the choice of the classifier when processing data should also depend on the chosen location and the target movement.

Mobile semantic Web provides a new way of developing context-aware social networking applications. Social networking applications are changing the way of communication by using userpsilas context-information. For example, micro-blogging has become a smart way of conveying the current situation and activity by using user context. There is currently a significant difference between using social networking applications on a static computer compared to a mobile device, even if current mobile devices are powerful and have good connectivity. The difference is primarily related to the mobility aspect since the user contexts may change more frequently and the user may not be able to interact with the mobile device. In this paper we identify common characteristics of current social networking applications and how they attract users. Finally, we propose an agent-based system architecture that is based on a distributed platform for developing semantic and semi-automated mobile social networking applications.