Automatic Annotation for Human Activity Recognition in Free Living Using a Smartphone

Abstract

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).

Enhanced Human Activity Recognition Based on Smartphone Sensor Data Using Hybrid Feature Selection Model

“…Figure 9 depicts the accuracy of individual activity with feature selectiona and without feature selectio. In [31], the researchers used neural networks and random forests to detect human activities. In previous research, accuracy rate was below than 95% for different activities on average.…”

Assistive Conversational Agent for Health Coaching: A Validation Study

“…18,19 To ease this burden while improving the care process, Ibrahim et al 20 developed a multiagent platform to automate the process of collecting user-provided clinical outcome measures without clinician intervention. Health coaching approaches are widely adopted in various health domains to monitor cardiac rehabilitation, 21 promote physical activity at home for elderly, 22 medication adherence, 16 assist pregnant women, 23,24 promote healthy diet, [12][13][14] support individuals with spinal cord injury, 25 and assist with hand therapy. 26 Health coaching systems may vary in their techniques to tackle health issues.…”

“…The role of mobile apps to facilitate behavior change has shown promising results in providing rich context information including an objective assessment of physical activity level and information on the emotional and physiological state of the person. 15,25 Current health coaching systems integrate artificial intelligence (AI)-based conversational agents powered with machine learning and natural language understanding. 23,31,32 Such systems are flexible and can respond to users' requests.…”

Collecting Labels for Rare Anomalies via Direct Human Feedback—An Industrial Application Study

“…Among the most typical application fields are medical applications like smoking detection [67], sleep detection [68], or affect recognition [69] and the large field of activity recognition [70,71]. While earlier work has focused on collecting labels from diaries filled out by study participants, smartphone apps have taken over the field of human annotation [72][73][74][75]. The main advantage of collecting labels via smart phones is timely labeling triggered by events (e.g., from sensor data) paired with visualization of context data in order to give the user a sensible amount of information during annotation.…”