Resumen de Digitalization capacity for knowledge acquisition: learning from health monitoring
Digital transformation boosts the integration of intelligent data into all areas of society, from personal streams of life-span, an organization business workflows, to the whole ecosystem of different industries. The continuous connectivity and interaction in the digital world pave the way to learn knowledge from the variety of massive data. The Internet of Things (IoT) is a promising practice in the digitalization process. Its basic spirit is to thrust a paradigm that everything (machine and people) can be seamlessly connected into an IoT network by sensors.
Toward the next frontier society 5.0, it is aimed for a prosperous human-centered society where people can have a high quality of life. However, general IoT architectures and data value chain models are still device/platform-specific, which lacks necessary emphasis on people dimension. Health as the core aspect of people has a significant impact on their quality of life. The adverse health factors may cross the entire lifespan: from home to workplace, from commute to work or fitness, and even elderly people care homes. To center the research work, the research question was pinpointed: How to accelerate and enhance people’s health and well-being in the IoT data value chain? On the premise of research status-quo and the pinpointed research question, four specific research objectives were defined as follows. 1. Enhance people dimension, especially health and well-being perspective in conventional IoT architectures. 2. Develop continuous long term monitoring solutions to support better health management. 3. Learn psychological and physical health impact from (e.g., workers) a group of people over the activities. 4. Accelerate health-related data sharing with security trust and privacy assurance.
To fulfill the research objectives, the action research method was employed as a theoretical approach to analyze and implement the research; a guiding framework was designed to set up the research theme and context, the applied key concepts, enabling technologies, devices, and developed prototypes were introduced under the guiding framework; furthermore, several study designs were detailed presented including experimental setup, data collection, and relevant data analysis method.
Toward a human-centered society, the future of IoT is also seeking enhanced people-centered solutions. To accelerate and facilitate health-related data-driven knowledge acquisition and data value chain to society, organizations, and individuals, the research leverages advance technologies such as IoT, Smart Wearables, DLT, and machine learning techniques. To summarize, the study focuses on the health dimension, the thesis generalizes highly the main contributions. 1. The study proposed IoT application architectures such as Healthy operator 4.0 architecture and proved their feasibility through real-world application cases in the industry. 2. Three long-term monitoring solutions were developed using low-cost IoT devices and successfully adopted in practical usage to continuously collect health-related parameters. 3. Different data mining approaches and machine learning methods were investigated and compared to learn health impacts over the activities from a group of people. The method chosen was proved to be capable of better understanding people's behavioral patterns and hidden rules, by the real-world empirical analysis conducted in both Spain and the USA. 4. A data sharing solution was designed in the study, that integrates DLT (IOTA Tangle) to IoT data management, by which data transparency and data ownership can be implemented under a secure, fee-less, and trust data sharing mechanism.
The value produced by the contributions is reflected on the individual level, organization level, and society level, which lies in societal aspects such as smart environment, industry 4.0, and smart city. With data-driven AI technology booming, big data analytic era comes. The future of work is now. The advance technology such as deep learning, Hadoop, Kubernetes, and Spark can be employed to dig knowledge out of data. The IoT big data analytic can achieve an improved understanding of data for individuals, organizations, and society, to make efficient and effective decisions.
