Ensemble learning-based Interactive Machine Learning (IML) offers significant potential for enhancing user-adaptive healthcare monitoring in resource-constrained environments, where preserving privacy and minimizing interaction burden are crucial. This work presents a novel cascade strategy that uses discriminative models to integrate personalization into interactive learning pipelines. It enables lightweight, adaptive activity recognition systems suited for deployment in edge and home-based healthcare scenarios (i.e., local devices and in-home systems). To achieve this, the proposed cascade dynamically adjusts confidence thresholds to match user engagement, triggering selective interaction only when prediction certainty is low. User-labeled data is then used to incrementally retrain the models, enabling personalized adaptation over time. We evaluate the approach on a hydration-related activity dataset as an example of routines relevant to smart health monitoring. Results show that even limited user participation yields measurable improvements, with personalized models achieving higher accuracy than generic ones using only a few additional annotations. This demonstrates the feasibility of combining low-effort interaction with edge-compatible Artificial Intelligence (AI) to enhance real-time, personalized decision-making. Thus, the method supports the development of efficient, adaptive, and human-aware health solutions.

The adoption of IoT technology to improve wellness and health awareness in workplace environments is increasingly vital. However, the acceptance of such e-health interventions largely depends on employees' perceptions of their value versus the privacy and security risks associated with data usage. Hence, addressing these concerns is paramount for fostering trust and ensuring the successful integration of technology. For this reason, this work explores the critical role of privacy in the workplace and examines how concepts of data control and ownership can improve the acceptance and effectiveness of IoT solutions. Leveraging insights from an online questionnaire with 524 participants from European countries, the obtained findings contribute to the existing literature by expanding the understanding of data control in workplace contexts and providing valuable insights for designing IoT-mediated e-health interventions that effectively address employees' privacy concerns.

As the number of interconnected devices increases and more artificial intelligence (AI) applications upon the Internet of Things (IoT) start to flourish, so does the environmental cost of the computational resources needed to send and process all the generated data. Therefore, promoting the optimization of AI applications is a key factor for the sustainable development of IoT solutions. Paradigms such as Edge Computing are progressively proposed as a solution in the IoT field, becoming an alternative to delegate all the computation to the Cloud. However, bringing the computation to the local stage is limited by the resources’ availability of the devices hosted at the Edge of the network. For this reason, this work presents an approach that simplifies the complexity of supervised learning algorithms at the Edge. Specifically, it separates complex models into multiple simpler classifiers forming a cascade of discriminative models. The suitability of this proposal in a human activity recognition (HAR) context is assessed by comparing the performance of three different variations of this strategy. Furthermore, its computational cost is analyzed in several resource-constrained Edge devices in terms of processing time. The experimental results show the viability of this approach to outperform other ensemble methods, i.e., the Stacking technique. Moreover, it substantially reduces the computational cost of the classification tasks by more than 60% without a significant accuracy loss (around 3.5%). This highlights the potential of this strategy to reduce resource and energy requirements in IoT architectures and promote more efficient and sustainable classification solutions.