One of the most common issues among the older people is falling. Sensor technologies can control the risk of people to fall and make them aware of their health status. Numerous devices are available in the market and new brands are continuously deployed making challenging the choice of the ideal monitoring device. The present study aims at providing a methodology to identify older adults at risk of falls and guide the user in the choice of the best set of technology to control and prevent the fall risk. Multidimensional health data of 121 older adults were collected, pre-processed and fed to five machine learning classifiers to automatically identify people at fall risk. A positive fall risk classification triggers a knowledge-based algorithm to identify the best matching device according to five categories (type, accessory parameters, accessory functions, technical requirements, and digital effort) and considering the user’s clinical and social needs. This way, the algorithm is not constrained to a specific device but is able to handle several technologies and could be generalized to a larger pool of disease by properly training the ML classification models. Unlike common recommendation systems based on user preferences and user profiling, the proposed decision-support system does not require any a priori knowledge of the user preferences. Indeed, as a future development, it is planned to consider user’s opinion obtained from usability and user experience tests to optimize the personalization of the assignment and improve the user’s engagement.