Big Data Based Architecture of the ADL Recognition System

In the age of the Internet of Things (IoT), every motion in our daily life can be captured and modulated to a digital data with smart portable devices. These data are very valuable in the presentation of someone’s living status, and in analyzing its health condition. Based on that, suggestions by professionals can be given directionally to improve the living quality. An arising issue is that, in this way the amount of data in the filter is big and the velocity of data flow is high. The original structure is strainful in handling such a large amount of data or such detailed data. An urgent requirement is to build a structure to support big data in the aspect of grabbing, filtering, analysis, and presentation. The ADL Recognition System collects information from elderly people, analyzes their behaviors, and presents them in a visualized way to specific users. It aims to provide a better nursing service for elderly people and a convenience in assessing health condition or nursing level for nurses and doctors. My work is to design and implement a big data based architecture of the ADL Recognition System so that it can accept more users and data flowing in. Besides, the architecture will be expandable in computation and storage and adaptive to the scale of the real application environment. The necessity and methodology of importing big data based architecture will be justified in each process of data filtering from the aspects of technologies